UNITED STATES DISTRICT COURT DISTRICT OF MASSACHUSETTS
This Document Relates To:
Case No. 1:02 CV 12146 (Judge Gertner)
PLAINTIFFS’ MEMORANDUM OF LAW IN OPPOSITION TO DEFENDANTS’ MOTION TO PRECLUDE THE EXPERT OPINIONS OF SCOTT D. HAKALA, M. LAURENTIUS MARAIS, REINIER KRAAKMAN AND BERNARD BLACK
TABLE OF CONTENTS
PRELIMINARY STATEMENT .......... 1
ARGUMENT .......... 4
I. DR. HAKALA’S OPINIONS ARE ADMISSIBLE PURSUANT TO THE FEDERAL RULES, DAUBERT AND FIRST CIRCUIT LAW .......... 5
A. Dr. Hakala Implemented A Valid Peer-Reviewed Event Study Methodology .......... 8
B. Dr. Hakala Implemented the Methodology in a Reliable and Replicable Manner .......... 13
1. Dr. Hakala Properly Identified Days on Which Significant AOL-Specific News Entered the Market.......... 13
2. Defendants’ Other “Replication” Arguments Are Equally Unfounded.......... 18
C. Defendants’ Meritless Arguments Concerning Dr. Hakala’s Identification and Treatment of “Relevant” Events, “New” News, and Confounded Events Do Not Warrant Precluding His Opinions .......... 20
1. Dr. Hakala Applied An Accepted Damages Analysis Methodology And Properly Identified Relevant Disclosures .......... 21
2. Dr. Hakala Adequately Accounted for Any “Confounding” Events .......... 25
3. Dr. Hakala Only Identified “New” News as Material .......... 29
D. Dr. Hakala’s Calculation of an Analyst Proxy Results in a Reliable, Conservative Estimate of The Impact of Defendants’ Misrepresentations and Omissions .......... 31
II. MARAIS’S TESTIMONY IS ADMISSIBLE UNDER THE FEDERAL RULES AND DAUBERT.......... 35
III. PROFESSOR KRAAKMAN’S TESTIMONY IS ADMISSIBLE UNDER THE FEDERAL RULES AND DAUBERT .......... 37
A. Professor Kraakman is Amply Qualified to Serve as an Expert Concerning The Subjects of His Report.......... 37
B. Professor Kraakman’s Opinions are Relevant, Reliable and Will Assist the Jury In Making Important Factual Determinations In This Case.......... 39
C. Professor Kraakman’s Opinions Concern Issues Of Economic and Finance Theory and Do Not Constitute Legal Opinions .......... 40
D. Professor Kraakman’s Opinions Properly Rebut Professor Stulz’s Report and Testimony .......... 42
IV. PROFESSOR BLACK’S OPINIONS ARE ADMISSIBLE UNDER THE FEDERAL RULES AND DAUBERT .......... 43
A. Professor Black’s Extensive Corporate Finance Credentials And Experience Amply Qualify Him To Opine On The Significance Of Certain Statements To The Market .......... 44
B. Professor Black’s Report Properly Rebuts Opinions Offered By Professors Stulz And Deighton.......... 45
C. Professor Black’s Opinions Are The Product of Reliable Principles And Methods Which He Properly Applied To The Facts of This Case .......... 48
TABLE OF AUTHORITIES
Basic Inc. v. Levinson, 485 U.S. 224 (1988).......... 3, 38
Bonner v. ISP Tech., Inc., 259 F.3d 924 (8th Cir. 2001) .......... 5
Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579 (1993).......... 1, 4, 21
DeMarco v. Lehman Bros., Inc., 222 F.R.D. 243 (S.D.N.Y. 2004) .......... 33
Elwood v. Pina, 815 F.2d 173 (1st Cir. 1987).......... 37
Freeland v. Iridium World Comm’n, Ltd., 545 F. Supp. 2d 59 (D.D.C. 2008) .......... 49
Freeland v. Iridium World Commc’n, Ltd., 233 F.R.D. 40 (D.D.C. 2006).......... 29
In re AOL Time Warner, Inc. Sec. and “ERISA” Litig., No. MDL-1500 (S.D.N.Y.) .......... 36
In re Clarent Sec. Litig., No. 01-3361 CRB (N.D. Cal. Feb. 9, 2005) .......... 7
In re Countrywide Financial Corp. Sec. Litig., 588 F. Supp. 2d 1132 (C.D. Cal. 2008) .......... 28, 29
In re Daou Sys., Inc., 411 F.3d 1006, 1025 (9th Cir. 2005) .......... 26
In re Enron Corp. Sec., Derivative and ERISA Litig., No. MDL-1446, 2005 WL 3504860 (S.D. Tex. Dec. 22, 2005).......... 25
In re Flag Telecom Holdings, Ltd. Sec. Litig., 245 F.R.D. 147 (S.D.N.Y. 2007) .......... 8
In re JDS Uniphase Sec. Litig., No. 02-01486 CW (N.D. Cal. Nov. 1-2, 2007).......... 7
In re Motorola Secs. Litig., 505 F. Supp. 2d 501 (N.D. Ill. 2007) .......... 29
In re Nature’s Sunshine Prods. Inc. Sec. Litig., 251 F.R.D. 656 (D. Utah 2008) .......... 7
In re Neurontin Mktg., Sales Practices, and Prods. Liab. Litig., MDL No. 1629, Civil Action No. 04-10981-PBS, 2009 WL 1212944 (D. Mass. May 5, 2009) .......... 5
In re Omnicom Group, Inc. Sec. Litig., 541 F. Supp. 2d 546 (S.D.N.Y 2008) .......... 25
In re Omnicom Group, Inc., Sec. Litig., No. 02 Civ. 4483 (S.D.N.Y.) (Hr’g Tr. Aug. 24, 2007) .......... 18, 25
In re Parmalat Sec. Litig., No. 04-MD-1653(LAK), 2008 WL 3895539 (S.D.N.Y. Aug. 21, 2008) .......... 6, 7
In re PolyMedica Corp. Sec. Litig., 432 F.3d 1 (1st Cir. 2005).......... 3, 38
In re Raytheon Co. Securities Litigation, Civil Action No. 99-12142 (Saris, J.) (Order Denying Mot. to Strike Hakala Reports, May 6, 2004).......... 7
In re Salomon Analyst Metromedia Litig., No. 06-3225-cv, 2008 WL 4426412 (2d Cir. Sept. 30, 2008) .......... 31
In re Viagra Prods. Liab. Litig., 572 F. Supp. 2d 1071 (D. Minn. 2008) .......... 5
In re Vivendi Universal, S.A. Sec. Litig., No. 02 Civ. 5571 (RJH), 2009 U.S. Dist. LEXIS 34563 (S.D.N.Y. Apr. 6, 2009) .......... 29, 31
In re Xcelera.com Sec. Litig., 430 F.3d 503 (1st Cir. 2005).......... 6
In re Xcelera.com Sec. Litig., Civ. Action No. 00-11649-RWZ, 2008 U.S. Dist. LEXIS 77807 (D. Mass. Apr. 25, 2008) .......... 6, 7
In re Xcelera.com Sec. Litigation, Civ. Action No. 00-11649-RWZ, 2004 U.S. Dist. LEXIS 29064 (D. Mass. Sept. 30, 2004) .......... 6
Int’l Adhesive Coating Co. v. Bolton Emerson Int’l, Inc., 851 F.2d 540 (1st Cir. 1988).......... 50
Interfaith Cmty. Org. v. Honeywell Int’l, Inc., 399 F.3d 248 (3rd Cir. 2005) .......... 41
Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137 (1999).......... 1, 4, 5, 21
Lapin v. Goldman Sachs & Co., 254 F.R.D. 168 (S.D.N.Y. 2008) .......... 7
Lattanzio v. Deloitte & Touche LLP, 476 F.3d 147 (2nd Cir. 2007) .......... 26, 27
Lentell v. Merrill Lynch & Co., 396 F.3d 161 (2d Cir. 2005) .......... 25
Lormand v. US Unwired, Inc., No. 07-30106, 2009 WL 941505 (5th Cir. Apr. 9, 2009) .......... 25
McDonough v. City of Quincy, 452 F.3d 8 (1st Cir. 2006).......... 37
Nathenson v. Zonagen, 267 F.3d 400 (5th Cir. 2001) .......... 31
Nieves-Villanueva v. Soto-Rivera, 133 F.3d 92 (1st Cir. 1997).......... 41
Pelletier v. Main Street Textiles, LP, 470 F.3d 48 (1st Cir. 2006).......... 41
RMED Intern., Inc. v. Sloan's Supermarkets, Inc., No. 94 Civ. 5587 PKL RLE, 2000 WL 310352 (S.D.N.Y. Mar. 24, 2000) .......... 18
Robbins v. Koger Props., Inc., 116 F.3d 1441 (11th Cir.1997) .......... 26
Ruiz-Troche v. Pepsi Cola of Puerto Rico Bottling Co., 161 F.3d 77 (1st Cir. 1998).......... 4
Sappington v. Skyjack, Inc., 512 F.3d 440 (8th Cir. 2008) .......... 49
Shirk v. Fifth Third Bancorp, No. 05-cv-049, 2009 WL 692124 (S.D. Ohio Jan. 29, 2009) .......... 7
Tuli v. Brigham & Women’s Health Hospital, Inc., 592 F. Supp. 2d 208 (D. Mass 2009) .......... 40
U.S. v. Aviles-Colon, 536 F.3d 1 (1st Cir. 2008).......... 37
U.S. v. Buchanan, 964 F. Supp. 533 (D. Mass. 1997) .......... 41
U.S. v. Green, 405 F. Supp. 2d 104 (D. Mass. 2005) .......... 50
U.S. v. Hines, 55 F. Supp. 2d 62 (D. Mass 1999) .......... 39, 41
U.S. v. Vargas, 471 F.3d 255 (1st Cir. 2006).......... 49
Wagner v. Barrick Gold Corp., 251 F.R.D. 112 (S.D.N.Y. 2008) .......... 7, 8
Federal Rules of Evidence Rule 403 .......... 37
Rule 702 .......... 1, 4
Bradford Cornell & R. Gregory Morgan, Using Finance Theory to Measure Damages in Fraud on the Market Cases, 37 UCLA L. Rev. 883, 894 (1990) .......... 21
Confronting the New Challenges of Scientific Evidence, 108 Harv. L. Rev. 1532, 1552 (May 1995).......... 9
Frank J. Fabozzi et al., Financial Modeling of the Equity Market: From CAPM to Cointegration, 431-32 (Wiley 2006) .......... 18
Henry J. Cassidy, Using Econometrics: a Beginners Guide, (1981).......... 23
Information Asymmetry, the Internet, and Securities Offerings, 2 J. Small & Emerging Bus. L. 91-99 (1998) .......... 44
Intriligator, Econometric Models, Techniques, and Applications, 1978, pp. 188-189 .......... 23
J. Applied Corp. Fin. xxx-yyy (2000) (http://ssrn.com/abstract=84489) .......... 44
Joel E. Thompson, More Methods that Make Little Difference in Event Studies, 15(1) J. Bus. Fin. & Acct., 77, 78 (1988) .......... 10
John D. Jackson et al., The Impact of Non-Normality and Misspecification on Merger Event Studies, 13:2 Int. J. of the Econ. of Bus. 247, 262 (2006) .......... 2, 9
John Finnerty & George Pushner, An Improved Two-Trader Model for Estimating Damages in Securities Fraud Class Actions, 8 Stan. J.L. Bus. & Fin. 213, 219 (2003).......... 21, 22
John Y. Campbell et al., The Econometrics of Financial Markets, 524 (Princeton Univ. Press 1997).......... 14, 18
Litigation Services Handbook: The Role of the Financial Expert, 2005 Cumulative Supplement, 3d ed., John Wiley & Sons .......... 35
Madge S. Thorsen, Richard A. Kaplan & Scott Hakala, Rediscovering the Economics of Loss Causation, 6 J. Bus. & Sec. L. 93 (2006).......... 22
Michael Barclay & Frank C. Torchio, A Comparison of Trading Models Used for Calculating Aggregate Damages in Securities Litigation, 64(2)-(3) L. & Contemporary Probs. 105, 106 (2001).......... 21
Nihat Aktas et al., Event studies with a contaminated estimation period, 13 J. Corp. Fin., 129 (2007) .......... 2, 8
Peter Kennedy, A Guide to Econometrics, 253-54 (MIT Press 5th ed. 2003) .......... 14
Philip Hans Franses, Time Series Models for Business and Economic Forecasting, (Cambridge Univ. Press 2002) (1998) .......... 2
Richard Roll, R2, 43 J. Fin. 541, 558-560 (1988) .......... 9, 13, 14
Robert B. Thompson II, Chris Olsen & J. Richard Dietrich, The Influence of Estimation Period News Events on Standardized Market Model Prediction Errors, 63(3) Acct. Rev., 448, 466-68 (Jul. 1988).......... 10, 13, 14
Robert S. Pindyck & Daniel L. Rubinfeld, Econometric Models and Economic Forecasts, 162-166 (3d ed., McGraw-Hill Inc. 1991) .......... 23
Ronald J. Gilson & Bernard Black, The Law and Finance of Corporate Acquisitions, 221 (2d ed. Foundation Press, Inc. 1995) .......... 26
Self-Regulatory Organizations; Notice of Filing of Proposed Rule Change by the National Association of Securities Dealers, Inc. Relating to Issuer Disclosure of Material Information, 67 Fed. Reg. 142 (Aug. 2, 2002) .......... 14
 Lead Plaintiff Bricklayers and Trowel Trades International Pension Fund, on behalf of the certified class (“Plaintiffs”), respectfully submit this Memorandum In Opposition To Defendants’ Motion to Preclude the Expert Opinions of Scott D. Hakala, M. Laurentius Marais, Reinier Kraakman and Bernard Black, together with the Declaration of Melinda D. Rodon (the “Rodon Decl.”). As shown below, Plaintiffs’ expert witnesses are eminently qualified experts under Federal Rule of Evidence 702 and the Supreme Court’s holdings in Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579 (1993) and Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137 (1999), and should be allowed to offer their relevant and reliable opinions at trial.
In moving to preclude the testimony of all four of Plaintiffs’ expert witnesses, Defendants lodge myriad attacks attempting to paint Plaintiffs’ expert testimony as either irrelevant or unreliable, but a common theme reverberates throughout: the utter dearth of support, whether in academic literature or through any independent testing, for their arguments. Unable to even colorably impugn Dr. Hakala’s qualifications as an expert in the field of econometrics or the relevance of his opinions to factual issues in dispute, Defendants attempt to smear his credibility, applying factual determinations favorable to Defendants as a basis for arguing that Dr. Hakala’s methodology and professional judgments are unreliable, and claiming that Dr. Hakala must have employed a results-oriented approach to “manufacture” statistical significance for the disclosures at issue. Defendants’ unfounded and outrageous attacks grossly misconstrue Dr. Hakala’s proper  implementation of a valid methodology, and Defendants fail to submit to this Court any scientific study or regression analysis which demonstrates that an event study conforming to their criticisms would alter the results from Dr. Hakala’s study in any significant way. Indeed, as discussed in greater detail below, that is precisely what the Xcelera defendants did—they submitted an event study which they claimed corrected Dr. Hakala’s supposed errors and found an inefficient market with no statistically significant curative disclosures in that case. Here, to the contrary, Defendants and Professor Stulz concede market efficiency, and the only regression analysis Defendants have submitted confirms the statistical significance of the key disclosures in this case.
Thus, in contrast to Defendants’ baseless assertions and innuendo, the record before the Court amply supports the following facts which eviscerate Defendants’ arguments: (1) both Dr. Hakala’s proper event study and the biased event study by Professor Stulz clearly identify statistically significant price declines at the 90% confidence level or better associated with the major disclosure dates in this action – July 18-19, 2001, August 14, 2001, February 20, 2002, and July 24-25, 2002; (2) Dr. Hakala’s event study was conducted in accordance with peer reviewed literature and constitutes a more accurate measure of statistical significance than Defendants’ event study; and (3) all of Professor Stulz’s opinions regarding the supposed lack of causation and damages are not based on any scientific evidence but rather are circularly based on the assumptions and advice he received from defense counsel. (See Stulz Corrected Rpt. ¶  5(b) (“Dr. Hakala’s analysis and conclusions are severely flawed because they depend on several incorrect assumptions that counsel for CSFB has advised me are not supportable and contradicted by the facts developed in discovery and/or by reasonable economic analyses.”)) (Emphasis added.)
Finally, in Defendants’ desperation to eliminate Plaintiffs’ additional rebuttal experts, Defendants press such improbable arguments as claiming that Professor Reinier Kraakman, one of the foremost and oft-cited authorities on the efficient market hypothesis, is somehow inadequately qualified to opine regarding the economic implications of applying the efficient market hypothesis to analysts’ statements. (Defs.’ Mem. 44-46.) Likewise, they dubiously suggest that the testimony of Dr. M. Laurentius Marais, a published expert regarding event study methodology, is “irrelevant” because he rebuts only one of their several attacks on Plaintiffs’ event studies. (Defs.’ Mem. 32-33.) Defendants also challenge Professor Bernard Black, ignoring the majority of the opinions and analysis in his report and myopically and hypocritically arguing that a single, corrected error which did not significantly alter his ultimate conclusions somehow renders all of his opinions unreliable, despite the fact that Defendants’ own expert, Professor Stulz, had to correct four separate portions of his own report, including very significant and telling compensation information.
Hence, as explained in detail below, Defendants’ contrived and meritless arguments that the expert testimony of Dr. Hakala and Plaintiffs’ other experts is inadmissible should be rejected in their entirety.
Plaintiffs’ proffered expert testimony provides reliable specialized knowledge and information which is necessary to assist the jury with important factual determinations, and therefore is admissible under the Federal Rules of Evidence. Rule 702 of the Federal Rules of Evidence governs the admissibility of expert testimony. As the Supreme Court has explained, under Rule 702 the district court must perform a critical “gatekeeping” role to ensure that the scientific evidence admitted is both relevant and reliable. Daubert, 509 U.S. at 589. In performing this gatekeeping role, the district court enjoys “considerable leeway” in determining whether an expert’s testimony is reliable. Kumho Tire, 526 U.S. at 152. However, as the First Circuit has explained:
Daubert does not require that a party who proffers expert testimony carry the burden of proving to the judge that the expert's assessment of the situation is correct. As long as an expert's scientific testimony rests upon “good grounds, based on what is known,” Daubert, 509 U.S. at 590 (internal quotation marks omitted), it should be tested by the adversary process-competing expert testimony and active cross-examination-rather than excluded from jurors' scrutiny for fear that they will not grasp its complexities or satisfactorily weigh its inadequacies, see id. at 596.
Ruiz-Troche v. Pepsi Cola of Puerto Rico Bottling Co., 161 F.3d 77, 85 (1st Cir. 1998). Hence, “[i]f an expert’s testimony is within ‘the range where experts might reasonably differ,’ the jury, not the trial court, should be the one to ‘decide among the conflicting views of different experts.’” In re Neurontin Mktg., Sales Practices, and Prods. Liab. Litig., MDL No. 1629, Civil  Action No. 04-10981-PBS, 2009 WL 1212944, at *9 (D. Mass. May 5, 2009) (Saris, J.) (quoting Kumho Tire, 526 U.S. at 153). Indeed, “[o]nly if the expert's opinion is so fundamentally unsupported that it can offer no assistance to the jury must such testimony be excluded.” In re Viagra Prods. Liab. Litig., 572 F. Supp. 2d 1071, 1078 (D. Minn. 2008) (quoting Bonner v. ISP Tech., Inc., 259 F.3d 924, 929-30 (8th Cir. 2001)).
Defendants’ arguments fail to cast any doubt on the fact that Plaintiffs’ proffered expert testimony will assist the jury and is reliable and relevant. At best, Defendants’ arguments constitute disputes about which “experts might reasonably differ,” hence, Defendants’ motion to preclude Plaintiffs’ expert testimony should be denied in its entirety.
I. DR. HAKALA’S OPINIONS ARE ADMISSIBLE PURSUANT TO THE FEDERAL RULES, DAUBERT AND FIRST CIRCUIT LAW
Dr. Hakala performed his event studies in conformity with accepted practices, and his reports cite to numerous academic and peer-reviewed authorities that validate the methods he applied. (See Rodon Decl. Ex. A (Decl. of Scott Hakala Regarding Market Efficiency dated Feb. 28, 2007 (the “Market Efficiency Rpt.”)) ¶¶ 15-16 & nn.9-13; Rodon Decl. Ex. C (Expert Rpt. of Scott Hakala dated Mar. 4, 2008 (the “Damages Rpt.”)) ¶¶ 30-34.) Moreover, as discussed below, several peer-reviewed authorities published over the past two decades support the superiority of the event study performed by Dr. Hakala over the alternatives proposed by Defendants and Professor Stulz. Thus, using a proper methodology, Dr. Hakala’s event studies find statistical significance, evidencing loss causation and damages, associated with the materialization of the risks hidden by Defendants.
Despite the abundant support for Dr. Hakala’s analyses, Defendants attempt to label his methodology as unreliable, but fail to cite to any academic support or to submit any valid test or analysis girding their self-serving conclusions. Contrary to Defendants’ bald and counterfactual  assertions, Dr. Hakala’s approach was not outcome determinative, biased or otherwise unreliable. In attacking Dr. Hakala, Defendants substantially rely on In re Xcelera.com Sec. Litig., Civ. Action No. 00-11649-RWZ, 2008 U.S. Dist. LEXIS 77807 (D. Mass. Apr. 25, 2008) (Zobel, J.), which excluded an event study prepared by Dr. Hakala, notwithstanding the Xcelera court’s previous reliance on his event study at class certification and the First Circuit’s approving reference to that event study in upholding class certification. However, in Xcelera, the record before Judge Zobel was far different from the record here. Defendants in that case proffered an event study which concluded that the market was inefficient and that there were no statistically significant declines associated with any of the alleged disclosures of the misrepresented risk.
Thus while the plaintiffs in Xcelera may have disagreed with Judge Zobel’s finding in favor of defendants, that finding did have some basis in the record. Not so here. As set forth above, despite having had ample time to do so, Defendants simply have not and cannot produce a scientific study which finds no statistical significance associated with the major disclosures in this case. See In re Parmalat Sec. Litig., No. 04-MD-1653(LAK), 2008 WL 3895539, at *10 (S.D.N.Y. Aug. 21, 2008) (rejecting arguments against Dr. Hakala’s event study because opposing expert “did no event study to show how Dr. Hakala’s analysis would have differed had the claimed flaws in his work been corrected.”). Nor have Defendants presented a single academic test or paper that rejects or criticizes Dr. Hakala’s event study methodology. And indeed, Defendants have conceded that the market is efficient. Hence, at bottom, Defendants’ arguments against Dr. Hakala in this case are truly baseless, lacking any support in the record.
 Moreover, as discussed more fully in Plaintiffs’ memoranda submitted on May 9, 2008 and June 3, 2008, the Xcelera court’s exclusion of Dr. Hakala’s study was based primarily on the court’s disagreement with aspects of his loss causation analysis relating to delayed investor reaction to alleged leakage of corrective information, and Dr. Hakala’s heavy, but necessary, reliance in that case on internet bulletin board postings to identify disclosure events, due to the very limited news and analyst coverage of the subject company. These conditions are unique to that case, as were the rejected aspects of Dr. Hakala’s study there, compelling a different outcome here.
Indeed, Dr. Hakala’s event studies have been admitted in nearly all of the cases in which they have been presented, including in this district in In re Raytheon Co. Securities Litigation, Civil Action No. 99-12142 (Saris, J.) (Order Denying Mot. to Strike Hakala Reports, May 6, 2004), and most recently, in: Shirk v. Fifth Third Bancorp, No. 05-cv-049, 2009 WL 692124, *7 (S.D. Ohio Jan. 29, 2009); In re Nature’s Sunshine Prods. Inc. Sec. Litig., 251 F.R.D. 656 (D. Utah 2008); Lapin v. Goldman Sachs & Co., 254 F.R.D. 168 (S.D.N.Y. 2008); In re Parmalat, 2008 WL 3895539, at *9-10; Wagner v. Barrick Gold Corp., 251 F.R.D. 112, 119-20 (S.D.N.Y. 2008); In re JDS Uniphase Sec. Litig., No. 02-01486 CW (N.D. Cal. Nov. 1-2, 2007) (testified at trial as damages expert); In re Clarent Sec. Litig., No. 01-3361 CRB (N.D. Cal. Feb. 9, 2005) (testified at trial as damages expert). Dr. Hakala brings this considerable expertise in real-world applications of event study methodology to bear on the complex issues of analyst statements and  impact here, which will be invaluable to the jury in determining issues of materiality, loss causation and damages.
A. Dr. Hakala Implemented A Valid Peer-Reviewed Event Study Methodology
The validity of the basic methodology applied by Dr. Hakala (and Defendants’ expert, Professor Stulz), the “intervention” or “event parameter” methodology, is not challenged by Defendants. Moreover, as a general matter, event studies are a commonly accepted way of demonstrating a significant correlation between stock price movements and events or disclosures as evidence relevant to market efficiency, loss causation, materiality and damages. See, e.g., In re Flag Telecom Holdings, Ltd. Sec. Litig., 245 F.R.D. 147, 170 (S.D.N.Y. 2007) (noting that “numerous courts have held that an event study is a reliable method for determining market efficiency and the market's responsiveness to certain events or information,” and crediting an event study by Dr. Hakala over the defendants’ abbreviated Daubert challenge); Wagner, 251 F.R.D. at 119-20 (same, and likewise crediting an event study by Dr. Hakala concerning market efficiency).
Defendants’ main attack on Dr. Hakala’s event study methodology is their utterly baseless claim that Dr. Hakala’s event study is “result driven, internally inconsistent and scientifically unsound.” (Defs.’ Mem. 4-5.) But these unsupported attacks do not gain any strength from mere repetition. As described at length in Dr. Hakala’s reports, each and every step of the event study methodology applied in both of Dr. Hakala’s event studies in this litigation is supported by peer-reviewed literature. (See, e.g., Market Efficiency Rpt. ¶¶ 15-16; Rodon Decl. Ex. E (Rebuttal Rpt. of Scott D. Hakala dated July 17, 2008, the “Damages Rebuttal Rpt.”)) ¶¶ 38, 41-44.) More particularly, Dr. Hakala’s use of dummy variables to control for significant news released during the study period is supported by academic studies. See, e.g., Nihat Aktas et al., Event studies with a contaminated estimation period, 13 J. Corp. Fin. 129  (2007) (concluding that their study’s “results highlight the importance of explicitly controlling for unrelated events occurring during the estimation window, especially in the presence of event- induced increase in return volatility.”); John D. Jackson et al., The Impact of Non-Normality and Misspecification on Merger Event Studies, 13:2 Int. J. Econ. Bus. 247, 262 (2006) (recommending that researchers “document that they have included all relevant events, both related and [un]related, in their event study model” because failing to identify and control for significant events causes “the true significance level of the test [to be] misstated”). As Dr. Marais explained with regard to the Aktas study, “Aktas et al. state the problem that Dr. Hakala’s method aims to solve (that is, properly taking account of ‘unrelated [company-specific] events’ in the data used for estimation), and offer as a ‘natural solution’ to this problem essentially the method employed by Dr. Hakala (that is, selecting estimation data ‘free of such contaminating events’).” (Rodon Decl. Ex. F (the Rebuttal Decl. of M. Laurentius Marais, dated July 16, 2008, (the “Marais Rpt.”)) ¶ 17.)
Indeed, academic studies comparing event study methodologies that control for significant news releases, like Dr. Hakala’s, to methodologies that do not, like Professor Stulz’s, universally conclude that Dr. Hakala’s methodology results in a more powerful test, meaning that there is a greater “likelihood that [the] test will declare an association when there actually is an association.” See, e.g., Richard Roll, R2, 43 J. Fin. 541, 558-60 (1988) (finding an increase in the explanatory power of event studies that control for all company-specific news listed in the Wall Street Journal and Dow Jones news service, with the biggest improvements involving companies that experienced major events such as takeovers or disasters); Robert B. Thompson II, Chris Olsen & J. Richard Dietrich, The Influence of Estimation Period News Events on  Standardized Market Model Prediction Errors, 63(3) Acct. Rev. 448, 466-68 (Jul. 1988) (concluding “that conventional market model parameter estimates are biased relative to those derived from the news-conditional market model,” and that there is an “increase in power [which] appears to be due primarily to the inclusion of a broad set of firm-specific news events (i.e., those reported in the Wall Street Journal Index) in the model specification.”); and Joel E. Thompson, More Methods that Make Little Difference in Event Studies, 15(1) J. Bus. Fin. & Acct. 77, 78 (1988) (finding that “extraneous events may increase the variance of a firm’s returns resulting in a larger estimated variance and thereby decrease the power of the test.”).
Defendants do not challenge the conclusions of these studies, but they implicitly contradict them – without any academic foundation – in arguing that Dr. Hakala’s dummying out of returns for days with significant AOL-specific news artificially “inflates” Dr. Hakala’s results and renders them unreliable. Defendants cite to no literature, testing or other support in the record for their claim that controlling for significant events imparts any bias into the tests’ results, which is contrary to the academic literature cited above. Instead, Defendants rely wholly on the ipse dixit of Professor Stulz, who claims that Dr. Hakala’s use of dummy variables “unfailingly inflates the statistical significance of abnormal stock returns of the subject company.” (Defs’ Mem. 20; Stulz Corrected Rpt. ¶ 100.) Indeed, as Dr. Marais noted in his report, “Dr. Stulz conspicuously fails to identify any violation by Dr. Hakala of any specific, generally accepted mathematical or statistical principle whose violation would logically impart  the purported ‘bias’ to the Hakala results.” (Marais Rpt. ¶ 15.) Moreover, Professor Stulz apparently ignores the fact that Dr. Hakala’s second event study in this case, which employs more dummy variables than his first study, resulted in lower statistical significance findings for 23 days in the study, including 3 days when CSFB issued AOL reports. (Compare Market Efficiency Rpt. Ex. B and Damages Rpt. Ex. B-1.) Indeed, to demonstrate the invalidity of Professor Stulz’s criticism that Dr. Hakala’s supposed “overuse” of dummy variables “unfailingly inflates” his statistical significance findings, Dr. Hakala tested it by reducing the number of dummy variables used to include only statistically meaningful events, and found that the resulting statistical significance findings were not materially altered overall, with a minor increase in the statistical significance findings of most of the events of interest – the opposite of the expected result according to Professor Stulz. (Damages Rebuttal Rpt. ¶¶ 46, 49 & Exhibit B-1 Limited).) Thus, Defendants’ and Professor Stulz’s claims that Dr. Hakala’s methodology “unfailingly inflates” his results is not only unsupported, but provably wrong.
Similarly, Defendants and Professor Stulz offer no academic studies, independent testing or other support for their claim that any increased findings of statistical significance represent artificial inflation or bias as opposed to simply the more accurate findings of a more powerful study. As Dr. Marais explains, “[i]t is true that any fraud-related price movement will likely stand out more prominently against Dr. Hakala’s material-news-free baseline than against a baseline tainted by other, irrelevant but potentially price-moving news events. Dr. Stulz is  mistaken, however, in claiming that this consequence is a “bias”; rather, this is precisely—and properly—the purpose of Dr. Hakala’s procedure.” (Marais Rpt. ¶ 13 (emphasis in original).) Indeed, Dr. Hakala directly demonstrated in his rebuttal report on market efficiency that controlling for significant news releases increased the power of his event study relative to Professor Stulz’s study. Dr. Hakala replicated Professor Stulz’s regression analyses, applying dummy variables to control for the same significant news events as in Dr. Hakala’s event study. (Rodon Decl. Ex. B (Rebuttal Decl. of Scott D. Hakala dated June 27, 2007, (the “Market Efficiency Rebuttal”)) ¶ 13.) This analysis revealed that if Professor Stulz had used dummy variables to control for the same significant AOL-specific news as Dr. Hakala did, it would have reduced the standard error in Professor Stulz’s regressions from 2.52% to approximately 1.80%, a 28.4% improvement in the power of Professor Stulz’s event study. (Id.) Defendants do not offer any testing or other refutation of these findings.
Further, Defendants do not and cannot point to a single academic study suggesting that there is some finite maximum number of days which may be excluded from a study period (whether by use of dummy variables or otherwise) before an event study is rendered unreliable.Instead, Defendants cite to a handful of academic sources which do not discuss the use of dummy variables in the context of performing event studies to suggest that this evinces that dummy variables should not be used at all, and argue that the Aktas study cited by Dr. Hakala does not actually support his method because it only involved a small number of significant events that were dummied out. However, as Dr. Marais explains, “Aktas et al. state no explicit limitation of the scope of their comments to the case of only a small proportion of contaminating  events, and no such limitation is logically implied by their analysis of the issue.” (Marais Rpt. ¶ 18.) Indeed, the “contaminating news events” in the Aktas study were created by its authors and inserted into the study period to provide a hypothetical scenario for use in testing the effectiveness of various methods of controlling for contaminating news events: to read any significance into the number of fictional contaminating events the study’s authors created represents nothing more than a baseless effort to distinguish a plainly relevant and applicable academic precedent for Dr. Hakala’s methodology here. Moreover, Defendants’ argument ignores the academic articles cited in the Aktas paper and other articles cited herein which control for a larger number of company-specific news events, similar to Dr. Hakala. Hence, Defendants’ criticism of the number of dummy variables used by Dr. Hakala has no academic basis and is meritless.
B. Dr. Hakala Implemented the Methodology in a Reliable and Replicable Manner
1. Dr. Hakala Properly Identified Days on Which Significant AOL-Specific News Entered the Market
As described in considerable detail in Dr. Hakala’s reports, Dr. Hakala applied generally accepted protocols in identifying days when significant AOL-specific news was released, and tested to verify the significance of the news identified through this procedure. (See generally Market Efficiency Rpt. ¶¶ 12, 16-17; Damages Rpt. ¶¶ 30, 34-35.) As recommended by the  relevant academic literature, Dr. Hakala selected events on an a priori or “blind” basis, meaning that the events were selected based solely on the substance of the news conveyed and without knowledge of whether AOL’s stock price moved that day. (Market Efficiency Rpt. ¶ 12; Damages Rpt. ¶ 30.) As a foundation for the identification of significant or “material” information, Dr. Hakala and his staff applied the categories of material information contained in the NASD guidelines, as recognized by the Securities and Exchange Commission, together with third party news and analyst reports recognized as material in the academic literature. (Id.) Additionally, in order to verify the results of this process for identifying significant news, Dr. Hakala performed an “F-test,” which is “a conservative test for the statistical significance of a group of events or explanatory variables.” (Damages Rpt. ¶ 35.) “[T]he F-test for significance of the identified events suggested a confidence level in excess of 99.99%[.]” (Id.) Neither Professor Stulz nor Defendants have challenged Dr. Hakala’s F-test results, nor have they performed any tests contradicting these results.
Nonetheless, Defendants argue that Dr. Hakala’s identification of significant events to include in his event study is unreliable because it involves some measure of professional judgment, which they claim is impossible to replicate. (Defs.’ Mem. 24-25.) The sole support Defendants offer for their claim that Dr. Hakala’s process of event identification is not replicable  is a comparison of three separate event studies which Dr. Hakala prepared for different purposes relating to AOL stock price movements: one preliminary study in the class action case against AOL (the “AOL Study”), and the two performed in connection with this litigation. Defendants argue that the different number of events identified in these studies, standing alone, demonstrates that Dr. Hakala’s method cannot be replicated and is therefore, unscientific. This argument not only vastly overstates the significance of the changes between the reports, but wholly ignores the logical reasons for the changes, and generally misconstrues the scientific issue of replication.
First, Defendants’ argument counterfactually presupposes that each of these event studies involved a review of precisely the same set of source documents, but that Dr. Hakala somehow identified substantially more documents fitting within the news categories described in the protocol of his reports for his Damages Study than for his AOL Study. Dr. Hakala’s reports and testimony do not support this supposition. Although Dr. Hakala’s reports identified categories of news he sought to include in his studies (as discussed above) and listed several news sources he searched (i.e., Factiva, LexisNexis, Bloomberg, L.P., etc.), they did not specify his search terms or protocols for the first step of his process: gathering news for the subsequent manual review. As Dr. Hakala testified, the additional events identified in the later event studies were the result of three changes in this first step of his procedure: (i) more thorough database searches in gathering the documents to be reviewed, (ii) specific efforts to identify analyst disclosures which were relevant to the event studies in this case, but not the AOL Study and; (iii) access to additional news sources not previously available to Dr. Hakala. (Hakala 2008 Dep.  Tr. 194:10-195:17; see also Damages Rebuttal Rpt. ¶ 45.) Thus, the fact that additional events were included in the later studies based on a broader group of source documents reveals nothing about whether the application of the review protocols is replicable.
Second, Defendants do not show, through testing or otherwise, that any differences in the identification of events between the two reports in this case biased Dr. Hakala’s results or even impacted the results significantly. As Dr. Marais testified, whether there is a reasonable degree of replicability to a test in the field of statistics does not depend on “calculating percentages of dates that match between two implementations,” but rather, requires “tracing whatever range of variation happens to occur in the implementation of the event selection protocol . . . to its consequences for final opinions or . . . for ultimate conclusions[.]” (Rodon Decl. Ex. K (attaching transcript excerpts from the Aug. 13, 2008 deposition of M. Laurentius Marais, (“Marais Dep. Tr.”)) 87:2-89:23.) Indeed, if Defendants’ definition of replication were applied, most of the peer-reviewed papers on event study methods would be rejected since nearly all involve numerous judgments and assumptions as to the appropriate model and variables to employ. Hence, “replication,” as used by Defendants, does not and can not logically apply to the judgments and assumptions underlying the statistical model, but rather, as described by Dr. Marais, only applies to the ultimate conclusions obtained once those judgments and assumptions have been made.
Here, the results of Dr. Hakala’s Damages event study do not in any way undermine the conclusion of his Market Efficiency study, namely, that AOL stock trades in an efficient market. Moreover, the resulting statistical significance findings are substantially similar, and any  differences are inconsequential to the ultimate conclusions drawn from the studies. Finally, as explained supra p. 11, any changes in the statistical significance findings – both increases and the 23 decreases which Defendants do not mention – are the proper result of a better specified test, and therefore, do not suggest any bias or otherwise unreliable methodology.
Additionally, Defendants ignore as though it were immaterial the fact that the present matter involves different issues than the AOL litigation, and that an event study pertaining to different issues may properly identify different relevant disclosures. As Dr. Hakala’s declaration in the AOL litigation stated, his study there was preliminary, and he was asked to merely “identify instances, if any, of loss causation associated with the allegations” in that matter. Considering that the AOL litigation concerned AOL’s accounting fraud and misstatements and not issues regarding analyst statements, clearly different disclosures constituted relevant events between the two cases. Moreover, given that most of CSFB’s 35 AOL reports were reiterations, it is unlikely that Dr. Hakala would have identified many of them as otherwise significant news for which he would need to control.
Finally, Defendants’ argument that Dr. Hakala’s event study is rendered unreliable merely because some professional judgment was used in identifying material news ignores the fact that academic literature expressly recognizes that aspects of an event study, including the process of identifying events to include in the model, call for the exercise of informed  judgment. Dr. Hakala’s exercise of his informed judgment thus adheres to academic protocols and does not impart any unscientific or unreliable element to his event studies.
Likewise, by extension, Dr. Hakala’s professional judgment calls do not render his event studies inadmissible, because Daubert “does not preclude an expert from relying upon his experience and judgment in coming to conclusions.” (Rodon Decl. Ex. J (In re Omnicom Group, Inc., Sec. Litig., No. 02 Civ. 4483 (S.D.N.Y. Hr’g Aug. 24, 2007)) (denying motion to preclude Dr. Hakala’s testimony and event study based, in part, on Dr. Hakala’s selection of event and relevant days).) See also RMED Intern., Inc. v. Sloan's Supermarkets, Inc., No. 94 Civ. 5587
PKL RLE, 2000 WL 310352, *8 (S.D.N.Y. Mar. 24, 2000) (noting that “a statistical event study involves subjective elements. A researcher performing an event study must identify which company-specific events to study, and in the process, categorize those events as fraud or non- fraud related. . . . Because [the expert’s] decision was informed by a detailed factual analysis and grounded on principles generally accepted within the relevant field, her testimony is sufficiently reliable to be admitted.”) Hence, Defendants’ arguments that Dr. Hakala’s event studies should be precluded because they involved some professional judgment should be rejected.
2. Defendants’ Other “Replication” Arguments Are Equally Unfounded
Defendants also argue that changes Dr. Hakala made in the study period and composite index from the AOL Study to the Damages Report Study further demonstrate that his methodology cannot be replicated and is therefore unreliable. Like their arguments about Dr. Hakala’s identification of significant events, Defendants’ arguments regarding changes to Dr.  Hakala’s study period and composite index are based on an erroneous understanding of the concept of replication in statistical studies, and ignore the legitimate reasons connected to the facts of this case for the changes made. Because Defendants fail to establish any bias or impact on the results traceable to the changes of which they complain, there can be no doubt as to the replicability and reliability of Dr. Hakala’s event studies.
Defendants claim that Dr. Hakala “inexplicably” shortened the event study window from a September 16, 2002 ending date in the AOL Study to a July 25, 2002 ending date in the CSFB Damages Report Study, despite a comment in his AOL Study about the “reliability and precision” of his estimates there. (Defs.’ Mem. 26.) Defendants’ argument that this demonstrates a shortcoming in the study window of Dr. Hakala’s Damages Report strains credulity. In the first instance, there is nothing “inexplicabl[e]” about this difference between the two reports: Dr. Hakala expressly stated in his AOL Study that “I extended the study period beyond the July 26, 2002 date . . . in order to analyze the events that occurred between July 26 and September 16, 2002.” (AOL Study ¶ 30.) Defendants ignore that the AOL Study included disclosures after July 25, 2002 which were relevant to the loss causation analysis in that case, but which are not relevant to loss causation or damages here. Thus, Dr. Hakala’s use of a shorter event window here represents a proper limitation connected to the facts of this case.
Additionally, Defendants argue, without support of any kind, that changes in the composite index of Dr. Hakala’s AOL Study to the Damages Study were “arbitrary.” (Defs.’ Mem. 26.) However, as Dr. Hakala explained in his Damages Rebuttal report, the shorter study period here (January 12, 2001 through July 25, 2002) relative to the original study period in the AOL case (1999 through July 25, 2002) resulted in changes in the companies which constituted  significant competitors of AOL. (Damages Rebuttal ¶ 48 (regarding the rationale for the addition of Gannett and the removal of AT&T, Comcast and Ebay from the respective industry indices).) Moreover, Defendants do not explain why this change to fit the circumstances of the present matter should be considered “arbitrary” or how it might have harmed the results of Dr. Hakala’s Damages study. They do not explain this because they can not: the correction to the relevant competitors specified in Dr. Hakala’s market models produced a slight improvement in the explanatory power (meaning the percentage of the daily variance in AOL’s stock prices that was explained by the model), from 51.3% in the Market Efficiency study to 52.13% in the Damages Report study. (Market Efficiency Rpt. ¶ 13; Damages Rpt. ¶ 31.) Thus, the changes from the AOL Study are clearly based on a valid rationale and have the effect of improving the reliability of Dr. Hakala’s Damages Study for the purposes of this litigation.
C. Defendants’ Meritless Arguments Concerning Dr. Hakala’s Identification and Treatment of “Relevant” Events, “New” News, and Confounded Events Do Not Warrant Precluding His Opinions
Defendants further argue that Dr. Hakala’s opinions should be precluded as unreliable because he supposedly used irrelevant disclosures in his Damages Study, identified “old” news as material, and failed to take into consideration confounding events. (Defs.’ Mem. 8-12.) Defendants have not proffered an event study to show how any individual event Dr. Hakala supposedly erroneously included in his Damages event study biased or otherwise affected his results, but rather, aim to taint the Court’s opinion of Dr. Hakala’s judgments in general, offering only Defendants’ own severely flawed and self-serving factual assumptions as a basis. As described more fully below, Dr. Hakala’s judgments regarding the events to include in his study were reasonable and supported by the record. To the extent that Defendants raise any genuine  factual dispute as to the inclusion of particular “relevant,” “new” or “confounded” events in the Damages Study, these arguments easily fall within “the range where experts might reasonably differ,” Kumho Tire, 526 U.S. at 153, and thus, the “traditional and appropriate means of attacking” Dr. Hakala’s opinions on these matters is through “[v]igorous cross-examination, presentation of contrary evidence, and careful instruction on the burden of proof[.]” Daubert, 509 U.S. at 596. Thus, Defendants’ arguments fall far short of establishing any unreliability in Dr. Hakala’s opinions or analysis, as required for excluding his testimony under the Federal Rules of Evidence.
1. Dr. Hakala Applied An Accepted Damages Analysis Methodology And Properly Identified Relevant Disclosures
For the purpose of calculating damages, in accordance with the peer-reviewed literature Dr. Hakala sought to identify “the portion of the loss in share price . . . that would not have occurred had the truth as alleged by Plaintiffs been disclosed in a timely manner.” (Damages Rpt. ¶ 36.) To accomplish this, Dr. Hakala first established the total fraudulent inflation in the stock for each of the claimed misrepresentations and omissions, using the alleged curative disclosures for the accounting and layoff claims, and a calculation of the impact of comparable performance and ad-related analyst disclosures for the performance and ad-related claims, and  attributed that percentage of inflation to AOL’s share price as of the first relevant date for each claim (i.e., July 12, 2001 and January 12, 2001, respectively). (Id. at ¶¶ 9-11, Exs. C-1 & C-2. He next calculated AOL’s “true value” (i.e., its price per share excluding the fraudulent inflation) for each day from the beginning of the Class Period until all of the fraudulent inflation had dissipated from the stock through curative disclosures and other relevant disclosures concerning the misrepresented and omitted information. (Damages Rpt. ¶¶ 44-46, Exs. C-1 & C-2.) From this analysis, individual plaintiff’s damages can be calculated as the difference between the fraudulent inflation in AOL’s share price upon purchase less the inflation at the time of sale. (Damages Rpt. ¶¶ 47-49.) However, Defendants challenge the curative and relevant disclosures identified in Dr. Hakala’s analysis, claiming that they were selected in an improper manner and that certain of them are unrelated to the alleged fraud.
First, Defendants falsely and disingenuously claim that Dr. Hakala somehow “manufactured” the relevance of the dates in his Damages study by identifying the relevant events after running his regression analysis. (Defs.’ Mem. 14-15.) This bald claim is meritless, Dep. Tr. 224:23-225:18.) Based on this calculation, Dr. Hakala attributed only a small proportion of the total impact of each individual relevant event to Defendants. (See Damages Rpt. Exs. C-1 & C-1a (attributing only 13.41% of the relevant, performance and ad-related disclosures to CSFB).)  and indeed, undermined by the record. As discussed supra p. 14, Dr. Hakala selected all of the events included in the Damages event study blindly, i.e., based on the news imparted, not on the corresponding return data. He likewise selected the relevant events based on the content of the news disclosed and not the statistical significance of the events – a fact not lost on Defendants, who criticized Dr. Hakala, essentially, for not data-mining and removing relevant events without statistically significant returns. (Defs.’ Mem. 27.) Thus, Dr. Hakala plainly did not select relevant events in a manner that could even potentially impart any bias.
Moreover, Defendants do not demonstrate how any of Dr. Hakala’s determinations of relevant events supposedly impacted the results of his analysis, as they claim. As with their other claims of bias, they do not demonstrate it because they cannot. Given that Dr. Hakala only attributed a small proportion of the total impact of any relevant event to Defendants (based on his calibration calculation discussed, supra note 32), the impact of each individual relevant event is quite small on the amount of fraudulent inflation reflected in the damages analysis. (Damages Rpt. Ex. C-1.) Moreover, if anything, Dr. Hakala’s identification of relevant leakage events benefited Defendants by dissipating the fraudulent inflation out of the stock price earlier in the Class Period and thereby reducing damages overall: indeed, 32 of the 50 relevant events  identified by Dr. Hakala reduced the amount of inflation for shares purchased after those events. (Id.) Hence, Defendants fail to identify how Dr. Hakala’s process might potentially have biased his results, much less demonstrate any actual bias.
Defendants further argue that Dr. Hakala’s definition of a “relevant event” is unclear and that it is “impossible to discern how many of the ‘relevant events’ listed in Dr. Hakala’s exhibits have anything to do with this case.” (Defs.’ Mem. 18.) However, Defendants’ relevancy arguments are rooted in Defendants’ misguided belief that CSFB’s AOL reports are the only relevant disclosures in this litigation. For example, Defendants argue that Dr. Hakala’s failure to designate 22 of their 35 reports on AOL as relevant events, and his designation of February 1, 2001 and March 7, 2001 as relevant deflationary events due to disclosures other than the CSFB AOL reports issued on those days, renders his analysis in conflict with the allegation that CSFB’s AOL reports caused AOL’s share price to be artificially inflated. (Defs.’ Mem. 17-18.) This is, in essence, simply a rehashing of Defendants’ perennial favorite (but legally unsound) argument, that they cannot be held liable for securities fraud under Plaintiffs’ theory, that Defendants’ reports maintained inflation in AOL’s stock by misrepresenting and omitting risks known to them that ultimately materialized, causing Plaintiffs’ losses. It thus is Defendants, not Dr. Hakala, who ignore the facts and allegations in this matter, as well as a considerable body of cases providing that loss causation and damages may be established with  evidence of the materialization of risks or third party curative disclosures. See Lormand v. US Unwired, Inc., No. 07-30106, 2009 WL 941505 (5th Cir. Apr. 9, 2009) (acknowledging that “the great weight of federal courts . . . have held that Dura does not prevent a plaintiff from alleging or proving loss causation by showing partial or indirect disclosures of such truth by persons other than the defendants).
2. Dr. Hakala Adequately Accounted for Any “Confounding” Events
Defendants argue that Dr. Hakala’s Damages event study failed to disaggregate the effects of supposedly confounding news, and therefore, is unreliable. (Defs.’ Mem. 8-10.) However, at best, these factual arguments do not render Dr. Hakala’s event study unreliable or inadmissible, but merely bear on the weight of the loss causation and damages evidence. (See, e.g., Rodon Decl. Ex. J (In re Omnicom Group, Inc., Sec. Litig., No. 02 Civ. 4483 (S.D.N.Y. Hr’g Aug. 24, 2007)) (“Defendants’ argument that Dr. Hakala failed adequately to disaggregate the effects of fraud-related and non-fraud-related information on each relevant day . . . does not require exclusion” because “Rule 702, a rule of threshold admissibility, should not be transformed into a rule for imposing a more exacting standard of causality . . . simply because scientific issues are involved.”).)
 Moreover, considering that “[a] plaintiff is not required to show ‘that a misrepresentation was the sole reason for the investment's decline in value’ in order to establish loss causation,” In re Daou Sys., Inc., 411 F.3d 1006, 1025 (9th Cir. 2005) (emphasis in original), but need only calculate a “rough proportion of the whole loss” attributable to Defendants’ alleged misconduct, Lattanzio v. Deloitte & Touche LLP, 476 F.3d 147, 158 (2nd Cir. 2007), it is clear that Dr. Hakala’s analysis, which reasonably identifies confounding news and roughly apportions the impact of the news accordingly, will assist the jury in its determinations of loss causation and damages.
Defendants’ arguments that Dr. Hakala did not properly take into account any potentially confounding simultaneous news releases are undermined by peer-reviewed literature and the record in this case. The academic literature concerning event studies advises that “[a]n event study can tell us that something happened, but it can’t tell us why . . . . The event study technique does not eliminate the need to assess cause through deductive reasoning; it only – though this is substantial – helps delineate what needs to be explained.” Ronald J. Gilson & Bernard Black, The Law and Finance of Corporate Acquisitions, 221 (2d ed. Foundation Press, Inc. 1995).
Thus, Dr. Hakala reviewed the news on each day of his study to determine whether there were multiple pieces of significant news, and attributed only a proportion of the impact for particular days to account for other news. For example, Dr. Hakala determined that on September 19, 2001, there were similar significant, positive AOL comments issued by both CSFB and Bear Stearns, thus, Dr. Hakala attributed only 50% of the share price increase corresponding to this  news to Defendants. As Dr. Hakala explained, “we can draw an inference that when two analysts speak and speak contrary to market beliefs and expectations and give more than just a reiteration, that it does move the stock price[,] and CSFB was one of the two.” (Rodon Decl. Ex. K, (attaching transcript excerpts from the July 10, 2007 deposition of Scott D. Hakala, (“Hakala 2007 Dep. Tr.”)) 69:24-70:6.) Plainly, this analysis represents an exercise of deductive reasoning anticipated by the academic literature governing event studies. Moreover, it will assist the jury by delineating a “rough proportion of the whole loss” to Defendants’ alleged misconduct. Lattanzio, 476 F.3d at 158.
Additionally, the record establishes that for days when multiple analyst reports were issued, Dr. Hakala evaluated the reports and determined that certain of the days were not confounded (i.e., they did not contain different significant information) if: (i) one of the reports was clearly positive or negative and the other(s) was (or were) relatively neutral or (ii) both (or all) reports contained similarly positive or negative statements.” (Damages Rebuttal Rpt. ¶ 20.) Such un-confounded analyst event days were thus properly included in Dr. Hakala’s study. Additionally, as previously described, for damages purposes Dr. Hakala only ascribed a portion of the total impact of any such events to CSFB based on his calculation of the impact of CSFB’s misrepresentations concerning AOL’s performance and advertising revenues. Hence, Dr. Hakala  properly sought to identify any confounding news and ascribed only CSFB’s calculated proportion of the impact to Defendants.
Thus, contrary to Defendants’ claims that Dr. Hakala ignored the issuance of multiple “confounding” analyst reports, for example, in selecting February 1, 2001, July 19, 2001 and February 5, 2002 as relevant events for his study, Dr. Hakala in fact evaluated the reports to determine if they were confounded and discounted the impact of the news by 86.59% for each of these dates. (Id.) Defendants have provided no explanation of why this approach is unreasonable or how it renders Dr. Hakala’s analysis or results unreliable.
Finally, Defendants’ extreme position – that a plaintiff can never establish loss causation where there are multiple pieces of news or even just one piece of news which addresses multiple subjects – underestimates the jury’s capacity to review news and qualitatively weigh its significance with the assistance of expert testimony. Moreover, if the Court were to accept Defendants’ position, the result would be vast immunity against securities fraud liability for any company receiving adequate press or analyst coverage. Indeed, such a rule could even encourage companies to thwart liability by strategically releasing curative information coupled with other, unrelated information, as was recently acknowledged in In re Countrywide Financial Corp. Sec. Litig., 588 F. Supp. 2d 1132, 1200 (C.D. Cal. 2008) (observing that “corrective information often comes at the same time as good news (. . . either innocently or in order to minimize volatility or confound loss causation).”) Under the same reasoning with regard to claims under the Securities Act of 1933, the Countrywide court rejected defendants’ loss causation rule which “would perversely encourage slow information leaks and give management a strong incentive to correct market misperceptions as slowly and ambiguously as possible[,]” noting that “[i]f taken to its logical conclusion, Defendants’ rule would eliminate liability for  even the most egregious fraud where corrective disclosure comes in such minute increments that no plaintiff could locate a discrete point of ‘correction.’” Id. at 1172 n.49.
Thus, given Dr. Hakala’s reasonable efforts to identify and respond to confounding news, and the jury’s ability to weigh this information, Defendants’ arguments substituting an unduly strict loss causation standard for the relevant Rule 702 admissibility standard should be rejected.
3. Dr. Hakala Only Identified “New” News as Material
Defendants claim that “numerous” relevant events included in Dr. Hakala’s Damages study were not “new” news, and that his study therefore contradicts market efficiency and is unreliable. (Defs.’ Mem. 11-12.) However, like Defendants’ other self-serving factual findings, Defendants’ determinations that certain news was not “new” are severely flawed and do not constitute grounds for precluding Dr. Hakala’s testimony.
Defendants claim that their AOL report on February 5, 2001 is an example of “old” news which Dr. Hakala erroneously concluded impacted AOL’s stock, supporting this claim with nothing more than a misleading excerpt of a statement Dr. Hakala made at his deposition.
(Defs.’ Mem. 12 (quoting Dr. Hakala as saying, “[S]ometimes a report that might not change anything . . . can sometimes have a positive effect. And that’s an example of what’s happening here.”).) The complete statement uttered by Dr. Hakala was: “[S]ometimes a report that might not change anything but is reacting after a negative event such as occurred on February 1st, even reiterating a report that's following a negative development can sometimes have a positive effect. And that’s an example of what’s happening here.” (Rodon Decl. Ex. K (attaching  transcript excerpts from the Aug. 11, 2008 deposition of Scott D. Hakala (“Hakala 2008 Dep. Tr.”)) 214:12-18) (emphasis added).) Moreover, this statement was an explanation of how the qualitative statements in the February 5, 2001 CSFB report can impact a stock’s price, even when there are no quantitative changes to earnings forecasts, price targets and buy/sell ratings. Indeed, as Plaintiffs explained in the summary judgment briefing, Defendants’ February 5, 2001 AOL report clearly contained a new, significantly expanded optimistic commentary by Jamie Kiggen regarding AOL’s advertising revenues. (Plts.’ S.J. Opp. 64 (Dkt. Entry 288).) Thus, it was CSFB’s affirmative decision to promote AOL’s shares and reaffirm its confidence in AOL after doubts had been raised on February 1, 2001 that made the February 5, 2001 CSFB report material and significant to this case.
Likewise, Plaintiffs have thoroughly refuted Defendants’ simplistic conclusion that there was no new, relevant news in the February 20, 2002 Lehman Brothers report, which Dr. Hakala has also disputed. (See Plts.’ S.J. Opp. 57-58 (detailing the new information contained in Lehman’s February 20 report); Damages Rebuttal Rpt. ¶ 18 (stating that “the assertion . . . that the Lehman Brothers report on February 20, 2002 did not contain new information but merely repeated statements from a prior report on January 31, 2001 is simply not true, as a review of both reports easily reveals, and is belied by the stock price impact of the report and the numerous published news articles that attributed the decline in AOL’s share price to the February 20, 2002 Lehman Brothers analyst report.”).) Hence, Defendants’ unfounded and self-serving conclusions regarding “new” news do not suggest any true flaw in Dr. Hakala’s analysis, and should remain issues for the jury to resolve.
 D. Dr. Hakala’s Calculation of an Analyst Proxy Results in a Reliable, Conservative Estimate of The Impact of Defendants’ Misrepresentations and Omissions
Defendants criticize Dr. Hakala’s estimation of fraudulent inflation based on an analyst proxy, claiming that “under established law and economic principles,” he should have only analyzed the impact of CSFB’s reports, not the impact of other analysts’ reports. (Defs.’ Mem. 28.) This argument, again, rests on Defendants’ misguided belief that there can be no liability for their misrepresentations without proof of statistically significant stock price increases in response to their AOL reports. See In re Scientific-Atlanta, 571 F. Supp. 2d at 1340-41 (“Contrary to Defendants’ argument, the mere absence of a statistically significant increase in share price in response to fraudulent information does not ‘sever the link’ between the material misstatements and the price of the stock. Rather, price stability may just as likely demonstrate the market consequence of fraud where the alleged fraudulent statement conveys that the company has met market expectations, when in fact it has not.”). Considering that Defendants never issued anything even vaguely resembling a curative disclosure, Defendants’ argument that Dr. Hakala must rely solely on CSFB’s reports is nonsensical. Moreover, as detailed below, Defendants’ criticisms misstate or miscomprehend Dr. Hakala’s analysis, and should be rejected.
 In order to conservatively estimate the impact of an “equivalent disclosure” for the fraudulent inflation attributable to Defendants’ misrepresented performance projections and ad revenue risk, given that Defendants never timely or sufficiently lowered their AOL projections to match their internally discussed views, Dr. Hakala identified equivalent, significant negative analyst comments on AOL (i.e., significant ad revenue projection reductions, substantially reduced price targets or EBITDA estimates, downgrades, etc.), on days which were relatively free of AOL news from non-analyst sources, and found an average negative impact of 2.71% associated with such negative analyst forecast revisions. (Damages Rpt. ¶¶ 7 & 27; Hakala 2008 Dep. Tr. 187:15-189:16.) Dr. Hakala also considered a combined set of positive and negative AOL analyst reports issued on relatively news-free days, and found an average absolute impact of 2.70%. (Damages Rpt. ¶ 27.) Finally, Dr. Hakala compared these results to the actual inflationary impact of two significant CSFB AOL reports to confirm the validity of his estimate: one on February 5, 2001, which had a relative price effect of 2.91%, and the other on September 19, 2001, which had a relative price effect attributable to CSFB of approximately 1.79%, averaging out to 2.35%. Thus, Dr. Hakala used the estimated impact of 2.70% to represent the inflation in AOL’s share price on January 12, 2001 attributable to Defendants’ misrepresentations concerning AOL’s expected performance and ad revenues. during the Class Period. (Corrected Rpt. ¶ 38.) Considering that Plaintiffs allege that all of these reports misrepresented and omitted Defendants’ true beliefs about AOL’s poor financial prospects, it is hard to imagine how the allegedly fraudulent reports could possibly serve as a basis for estimating the impact of curative disclosures.
 Incredibly, Defendants argue that this analysis “has absolutely nothing to do with AOL’s actual stock price movements,” (Defs.’ Mem. 29), when Dr. Hakala’s entire analysis is plainly and precisely that, an analysis of AOL’s stock price responses to significant analyst forecast and valuation revisions comparable to the disclosure(s) Defendants should have made, given their internally expressed beliefs about AOL’s financial prospects.
Moreover, contrary to Defendants’ suggestion, Dr. Hakala’s analyst proxy bears no resemblance to the expert analysis rejected in DeMarco v. Lehman Bros., Inc., 222 F.R.D. 243, 248 (S.D.N.Y. 2004). (Defs.’ Mem. 29-30.) There, the expert had assumed a “minimum” analyst impact of 5% based on his review of a general study of the impact of analysts statements on 200 companies during the period of 1989-1991, and had estimated a “maximum” analyst impact based on the returns data for three days in the class period when the price of the subject stock rebounded after publication of analysts’ positive statements, yielding an average 16% maximum impact. DeMarco, 222 F.R.D. at 248-49. The DeMarco court found that this study was unreliable and irrelevant because, inter alia, the estimate of the “minimum” analyst impact was based on a study of analysts in general from a much earlier, pre-Internet bubble period, and the maximum estimate involved a very small sample of only three dates. Dr. Hakala’s study here is vastly different because, unlike the study in DeMarco, Dr. Hakala’s estimates: (1) are based on an event study which measured, rather than assumed, the impact of analyst statements; (2) used actual AOL return data from the Class Period, rather than return data for other companies from a different time period; (3) covered a much larger sample of 43 relatively news- free days, as opposed to the three days used in the DeMarco study; and (4) were confirmed by  comparison to two clean instances of inflationary CSFB reports on AOL. Thus, Defendants’ argument that Dr. Hakala’s study here is perilously similar to the DeMarco study is unfounded and should be rejected.
Finally, in response to the criticisms by Professor Stulz, including criticisms of a percentage-based damages calculation generally, Dr. Hakala has thoroughly explained the logical and economic underpinnings for his approach and provided extensive authority for his inflation per share methodology. (Damages Rpt. ¶¶ 33-48; Damages Rebuttal Rpt. ¶¶ 50-52.) Further, Dr. Hakala has submitted with his Damages Rebuttal an alternative analysis under a dollar-drop methodology which does not involve any analyst proxy analysis, and confirms the fact that there are substantial damages in this case under either methodology. (Damages Rebuttal Rpt. ¶ , Exs. C-1a (Limited Rebuttal) & C-2a (Limited Rebuttal).) Defendants have never offered any refutation of this analysis, and indeed, Professor Stulz argues that a dollar-drop method is the appropriate measure of damages in this case. (Corrected Rpt. ¶¶ 89-91 (arguing that a percentage approach is incorrect because, supposedly, “the same piece of information will affect the present value of the cash flows by the same dollar amount regardless of the current price of the stock at the time the market learns that piece of information[.]”).) Given Dr. Hakala’s presentation of both alternatives, which both find significant damages, and Defendants’ failure to offer any damages analysis, there simply is no basis in the record for Defendants’ claims that there is any flaw in Dr. Hakala’s damages analyses or that any such claimed flaw significantly undermines his results.
Thus, as described above, Dr. Hakala’s event studies and opinions are relevant, reliable, and will greatly assist the jury in resolving these complex issues, and therefore should be admitted under Federal Rule of Evidence 702.
 II. MARAIS’S TESTIMONY IS ADMISSIBLE UNDER THE FEDERAL RULES AND DAUBERT
Due to the numerous, complicated but technically meritless attacks lodged by Professor Stulz and Defendants against the event study methodology employed by Plaintiffs’ loss causation and damages expert, Dr. Hakala, Plaintiffs retained Dr. M. Laurentius Marais, a preeminent authority on event study methodology, to set the record straight on these significant issues. Dr. Marais is a coauthor of “Event study methods: detecting and measuring the security price effects of disclosures and interventions,”53 and has served on the editorial board of the Journal of Accounting Research and refereed for numerous professional journals, including the Journal of Accounting and Economics; the Journal of Business and Economic Statistics and the Journal of Financial Research. (Marais Rpt. Ex. A.) Dr. Marais reviewed Dr. Hakala’s event study methodology and opined unequivocally that it is consistent with peer-reviewed literature “is affirmatively and positively consistent with ideas that are in the peer reviewed literature.” (Marais Dep. Tr. 108:7-110:23.)
Defendants acknowledge that Dr. Marais answers Professor Stulz’s methodological criticisms, but claim that his testimony is irrelevant because Dr. Marais only opined on the validity of Dr. Hakala’s method and not various details of how the method was implemented. (Defs.’ Mem. 32-33.) This argument carries no practical import: clearly Dr. Hakala’s opinions may be excluded by the Court or discounted by the jury if they are determined to be based on an invalid methodology, irrespective of how that methodology was applied. Moreover, because  Defendants’ attacks on Dr. Hakala’s methodology confuse or ignore significant principles of statistics and misrepresent the academic literature supporting Dr. Hakala’s event study – literature which is completely indecipherable to a lay person lacking high-level statistics training – Dr. Marais’s testimony will assist a jury in understanding the key concepts and foundation for Dr. Hakala’s work here, which provides substantial evidence of loss causation and damages.
Defendants also argue that Dr. Marais’s opinions are redundant of Dr. Hakala’s defense of his methodology, and are therefore excludable as “needlessly cumulative” evidence. To the contrary, Dr. Marais’s testimony is not a mere reiteration of Dr. Hakala’s: Dr. Marais brings his considerable knowledge of the peer-reviewed academic literature regarding event studies to bear on the question of whether and how Dr. Hakala’s methodology is supported in the literature, offering in particular, an expanded discussion of the primary article Dr. Hakala cites for support, the Aktas article. As previously noted, Dr. Marais also offers unique testimony from Dr. Hakala on the issue of “replicability” in statistical studies, and concludes from his own review of Dr. Hakala’s described methodology that there was “no reason to expect” any increase rate of “false positive” results based on Dr. Hakala’s use of dummy variables. (Marais Dep. Tr. 167:2-168:4.) Hence, Dr. Marais’s testimony regarding Dr. Hakala’s methodology plainly exceeds the scope of Dr. Hakala’s own defense. Indeed, similar testimony by Dr. Marais supporting Dr. Hakala’s methodology was admitted in In re AOL Time Warner, Inc. Sec. and “ERISA” Litig., No. MDL-1500 (S.D.N.Y.). (See Market Efficiency Rebuttal ¶ 11.)
Finally, even if Dr. Marais’s testimony were cumulative in some sense of portions of Dr. Hakala’s testimony, this alone would not render it unduly or prejudicially cumulative under  Federal Rule of Evidence 403. See U.S. v. Aviles-Colon, 536 F.3d 1, 22-23 (1st Cir. 2008) (rejecting claim that witness’s testimony and recorded conversation presented at trial “were cumulative or unfairly prejudicial under Federal Rule of Evidence 403 simply because [another witness] had previously testified” as to the same matters). Further, the cases cited by Defendants do not support precluding a witness from testifying altogether based solely on the purportedly “cumulative” nature of the testimony offered. (See Defs.’ Mem. 33 (citing McDonough v. City of Quincy, 452 F.3d 8, 20 (1st Cir. 2006) (ruling that the district court did not abuse its discretion in urging the City to “move on” at trial after counsel had asked the plaintiff several repetitive questions about the plaintiff’s sour relationship with the City); Elwood v. Pina, 815 F.2d 173, 178 (1st Cir. 1987) (ruling merely that the exclusion of certain cumulative documentary evidence, the substance of which had been presented through testimony, “[i]f an error at all . . . was harmless.”).) Hence, Dr. Marais’s proposed testimony is neither irrelevant nor needlessly cumulative and Defendants’ motion to preclude it should be denied.
III. PROFESSOR KRAAKMAN’S TESTIMONY IS ADMISSIBLE UNDER THE FEDERAL RULES AND DAUBERT
A. Professor Kraakman is Amply Qualified to Serve as an Expert Concerning The Subjects of His Report
Without question, Professor Kraakman is qualified to testify as an expert on the issues addressed in his report and deposition, namely, matters of corporate finance and economic theory. (See Rodon Decl. Ex. G (the Expert Rebuttal Report of Reinier Kraakman to the Expert Report of Professor Stulz, dated July 16, 2008 (the “Kraakman Rpt.”)) ¶ 1 & Ex. A.) He is a highly respected academic who authored certain of the seminal articles concerning the functioning of efficient markets and the efficient market hypothesis. In fact, his publications have been cited in at least fifty court opinions, including most notably here, the Supreme Court’s decision in Basic and the First Circuit’s opinion in PolyMedica. See Basic Inc. v. Levinson, 485  U.S. 224, 253 (1988); In re PolyMedica Corp. Sec. Litig., 432 F.3d 1, 9 (1st Cir. 2005). Professor Kraakman has authored numerous textbooks and over forty articles on corporate finance and corporate governance, and currently teaches courses in corporations, corporate finance and corporate theory at Harvard Law. (Kraakman Rpt. Ex. A; Rodon Decl. Ex. K (attaching transcript excerpts from the Aug. 7, 2008 deposition of Reinier Kraakman, (“Kraakman Dep. Tr.”)) 14:14-24.)
Defendants nonetheless argue that Professor Kraakman lacks sufficient “‘knowledge, skill, experience training or education’ regarding the influence of analyst reports and other information on stock prices” based on Professor Kraakman’s testimony that he was “[p]robably not” more of an expert than defense counsel academic literature pertaining to analyst’s impact. (Defs.’ Mem. 42.) This argument strains credulity. Professor Kraakman’s modest self- assessment – or generous assessment of defense counsel’s knowledge – does not nullify the considerable expertise he has built in his field. Moreover, the issue is whether he has specialized knowledge which will assist the jury, and as Professor Kraakman further testified, “I think I’m more qualified to discuss the content of that literature than someone who hasn’t read the articles I’ve read.” (Kraakman Dep. Tr. 51:15-20.) Professor Kraakman noted that he had researched the literature concerning the impact of analysts’ statements in connection with his 1984 article, Mechanisms of Market Efficiency, and that he reviewed additional literature on the issue in preparation for his testimony in this case. (Id. at 50:13-51:6.) Thus, there is no genuine issue concerning Professor Kraakman’s impressive qualifications to opine as an expert regarding the efficient market hypothesis and its application to analyst statements.
 B. Professor Kraakman’s Opinions are Relevant, Reliable and Will Assist the Jury In Making Important Factual Determinations In This Case
Professor Kraakman opines here on whether and under what circumstances stock prices in an efficient market respond to information, including statements issued by analysts and others. These opinions will assist the jury in determining whether AOL’s stock was impacted by Defendants’ statements, and in sorting through the numerous arguments Defendants have asserted regarding whether certain news releases in Dr. Hakala’s event study are “new” news or are “confounded” by other simultaneous releases. Defendants’ assertions that Professor Kraakman opines on legal issues and that his opinions are unreliable are meritless and should be rejected.
Professor Kraakman’s guidance concerning the relevant criteria for the jury to consider in determining whether a statement impacted a stock’s price will assist jurors because the operation of efficient markets is a specialized topic which is “beyond the ken of the lay jury.” See U.S. v. Hines, 55 F. Supp. 2d 62, 64 (D. Mass. 1999) (Gertner, J.). Moreover, a basic understanding of how efficient markets operate is critical to determining key issues in this case, including whether Plaintiffs have established loss causation and how to calculate damages based on the curative disclosures identified in the complaint. Professor Kraakman’s report provides jurors with an explanation of the factors identified in the great body of academic literature dedicated to studying the types of disclosures that impact stock prices. Rather than evaluating all of the disclosures Professor Stulz opines on or Defendants’ raise in their arguments and instructing the jury on how they should resolve the disputes, Professor Kraakman explains how prominent  analysts’ statements and other news impact the price of widely-traded stocks, much like issuers impact stock prices when they make public disclosures. (Kraakman Rpt. ¶¶ 8-9.) Thus, Professor Kraakman provides the jury with “a context for considering the evidence before it, as opposed to a roadmap to a particular outcome.” See Tuli v. Brigham & Women’s Health Hospital, Inc., 592 F. Supp. 2d 208, 211 (D. Mass 2009) (Gertner, J.). His explanations as to how efficient markets operate will thus aid jurors in their understanding of the issues without “simply telling jurors what the outcome should be.” Id.
Moreover, Professor Kraakman’s opinions are based on reliable data, including the academic literature concerning the efficient market hypothesis, which contains peer-reviewed studies on the impact of different types of news and speakers on stock market prices. Additionally, to the extent he offered any opinions regarding discreet factual issues, he directly reviewed source documents, including documents produced in this case concerning Mr. Kiggen’s and Ms. Martin’s rankings and other indicia of elite analysts, and the January 31, 2002 and February 20, 2002 Lehman Brothers analyst reports, and applied the criteria discussed in academic literature to the documents and information he reviewed. (See Kraakman Rpt. ¶ 4.)
C. Professor Kraakman’s Opinions Concern Issues Of Economic and Finance Theory and Do Not Constitute Legal Opinions
Defendants argue that Professor Kraakman’s report amounts to little more than a legal opinion concerning whether the fraud-on-the-market presumption should apply in this case. (See Defs.’ Mem. 43.) However, the fact that Professor Kraakman is, among other things, a law professor does not convert every opinion he provides into a legal opinion. Although Professor Kraakman’s opinions here bear some significance to the resolution of legal issues in this case – and they would not be relevant if they did not – his opinions concern an economic and corporate finance issue, namely, the market impact of certain types of statements as discussed in academic  literature, not the case law. As described above, he is easily qualified as an expert in corporate finance literature and he acts in that capacity here, explaining relevant factors identified in the academic literature for determining whether certain disclosures economically impacted a stock’s price.
Indeed, Defendants’ unfounded assertion is belied by the contents of Professor Kraakman’s report. In his report, Professor Kraakman does not discuss any case law. (See id.) Moreover, the academic literature Professor Kraakman cites addresses economic issues, including: the content of financial analysts’ forecasts of earnings, the investment value of brokerage analysts’ recommendations, the performance of buy and sell recommendations, analysts’ predictive abilities, and investor reaction to celebrity analysts. (See Kraakman Rpt. nn.4-9). Professor Kraakman does not recite the holding of a single court case, draw legal conclusions, interpret the meaning of statutes or regulations, instruct the jury as to the applicable law, or usurp the role of the jury in applying the law to the facts before it. See Hines, 55 F. Supp. 2d at 72 (Expert testimony does not usurp the function of the jury where “[a]ll that the expert does is provide the jury with more information with which the jury can then make a more informed decision.”). Thus, Professor Kraakman clearly does not opine on an issue of law, and Defendants’ argument that his testimony should be excluded on this basis should be rejected.
D. Professor Kraakman’s Opinions Properly Rebut Professor Stulz’s Report and Testimony
Professor Kraakman’s report broadly rebuts Professor Stulz’s opinions about whether certain news is “new” or confounded by providing guidance to the jury about relevant considerations in making these factual determinations. Additionally, as concrete and relevant applications of the principles set forth in the academic literature he discusses, Professor Kraakman directly rebuts Professor Stulz’s opinions concerning: (i) the elite status of Mr. Kiggen and Ms. Martin and; (ii) whether the ad-related information in the February 20, 2002 Lehman report constituted “new” news. As Professor Kraakman explains, by examining “press coverage, professional prominence and affiliation with a prominent investment bank,” one can infer whether an analyst belongs to the “influential professional elite.” (Kraakman Rpt. ¶ 6.) Based on these characteristics, Mr. Kiggen and Ms. Martin qualify as influential analysts who could impact the stock prices of the companies they covered. (Id. at ¶ 7.)
Defendants grossly misrepresent Professor Kraakman’s opinions regarding the February 20, 2002 Lehman report, claiming that he agreed with Professor Stulz that the “the negative information about online advertising at AOL in the February 20, 2002 report had already been disclosed in Lehman Brothers’ January 31, 2002 report.” (Defs’ Mem. 50.) As Professor Kraakman explained in his report, ”[t]he February 20th [Lehman Brothers’] report contained a distinctly bleaker and more detailed discussion of AOL’s advertising prospects” than the January 31st report and, more importantly, “the February 20th report cast AOL’s advertising difficulties as a principal basis for lowering Lehman’s rating of AOL Time Warner.” (Kraakman Rpt. ¶ 17.) The changed stock recommendation in the February 20th report, from “2 Buy” to “3 Market Perform,” represented new information concerning the impact of the ad market on AOL’s future  prospects. (See id.) Thus, to claim that Professor Kraakman agreed with Professor Stulz that this was not new ad-related news flatly misstates Professor Kraakman’s opinion.
For all of the reasons herein, Defendants’ arguments to preclude Professor Kraakman’s opinions should be rejected.
IV. PROFESSOR BLACK’S OPINIONS ARE ADMISSIBLE UNDER THE FEDERAL RULES AND DAUBERT
Professor Black opines in his rebuttal report that: 1) the market was not already aware in 2001 of the scope of the advertising decline and its impact on AOL; 2) Mr. Kiggen and Ms. Martin were highly respected analysts whose statements would have affected AOL’s share price; 3) disclosure of CSFB’s internal views regarding the impact on AOL of the advertising slowdown would have substantially affected investor views about AOL due to the significance of AOL’s advertising revenues to the enormous growth rate applied in valuing the company; 4) the mere fact of an additional investigation into AOL’s accounting could call into question the integrity of AOL’s financial reporting generally, and thus could affect the value of its shares; 5) substantial, undisclosed layoffs at AOL would be highly relevant to investors; and 6) certain CSFB reports were not plausible when written and were not appropriately revised when they turned out to be overly optimistic.
Of all of these opinions, Defendants only directly challenge the opinion that CSFB’s reports were highly optimistic, claiming that his entire opinion is rendered unreliable by a single error in only one of the underlying analyses, which Professor Black acknowledged and corrected at his deposition. Due to this correction, Professor Black moderated the language of his opinion, stating that CSFB’s AOL projections were “highly optimistic” instead of “absurd,” but did not withdraw his opinion. (See Rodon Decl. Ex. K, (attaching transcript excerpts from the Aug. 19, 2008 deposition of Bernard S. Black (“Black Dep. Tr.”)) 104:2-5.) Therefore, there simply is no  basis for Defendants’ argument that the corrected opinion or analysis is unreliable. (Black Dep. Ex. 1, p. 9.)
Defendants otherwise move to preclude Professor Black’s opinions based only generally on the claims that Professor Black is unqualified to opine concerning the types of information that are important to investors and that his opinions exceed the scope of the opinions by Professors Stulz and Deighton which he is rebutting. As discussed below, these arguments are meritless, and hence, Professor Black’s opinions should be admitted.
A. Professor Black’s Extensive Corporate Finance Credentials And Experience Amply Qualify Him To Opine On The Significance Of Certain Statements To The Market
Professor Black’s considerable experience and academic work in the field of corporate finance spanning over 25 years provides a more than sufficient basis for his opinions here. As described at length in his curriculum vitae, Professor Black has experience in a broad range of corporate finance matters as an author, advisor, counsel, board member and professor. (Rodon Decl. Ex. H (the Expert Rebuttal Report of Bernard S. Black, dated July 17, 2008 (the “Black Rpt.”)) App. A.) Professor Black has served as a policy advisor to foreign countries concerning securities markets and has advised the Securities and Exchange Commission on various topics including proxy reform and corporate governance issues. (Id.) Indeed, Professor Black served as counsel to Commissioner Joseph A. Grundfest at the Securities Exchange Commission. Further, Professor Black taught at Stanford and Columbia, and has published textbooks and scholarly articles on corporate finance. (See id.) Currently, Professor Black holds a joint position as professor of finance at the McCombs School of Business and professor of law at the University of Texas Law School. (Id.; Black Dep. Tr. 16:4-7.)
 Defendants nonetheless seek to disqualify Professor Black, arguing that he may not have “any more expertise” than the jury in determining the information that would be significant to investors because Professor Black “has never been a securities analyst” and is “only” generally familiar with the academic literature concerning the impact of analyst statements. (Defs.’ Mem. 39.) This argument strains credulity. Professor Black testified that he is “knowledgeable in finance and in the valuation of companies” and that he has performed the types of calculations used in his report numerous times over the course of his career. (Black Dep. Tr. 15:12-16:10.) Moreover, Professor Black has expert “knowledge about what it is that investment banks do, that analysts do, [and] what kinds of information investors are looking for from analysts,” (Black Dep. Tr. 16:21-17:6), which he developed from his extensive research and writing about corporate acquisitions, experience on the boards of directors of various public companies, experience as an attorney in transactional mergers and acquisitions practice at a major law firm, and a experience as an expert witness in securities cases of various kinds. (Black Dep. Tr. 17:7-18:5.) Thus, there is no merit to Defendants’ argument that Professor Black is unqualified to opine on issues concerning company valuation and investors’ concerns merely because he has never worked as an equity research analyst.
B. Professor Black’s Report Properly Rebuts Opinions Offered By Professors Stulz And Deighton
Contrary to Defendants’ claims that Professor Black’s opinions fail to rebut the opinions of Professors Stulz and Deighton and should therefore be precluded, Professor Black’s entire analysis rebuts their opinions and criticizes their focus on irrelevant issues. As discussed above, Professor Black counters Professor Stulz’s analysis concerning the “reasonableness” of Defendants’ projections and criticizes Professor Stulz’s opinion that CSFB’s $75 price target could be justified. (See Black Rpt. 22.) Moreover, Professor Black rejects Professor Stulz’s  contentions, based on facts Professor Stulz was instructed by defense counsel to assume, that Defendants’ AOL reports were honest and reasonable, that Mr. Kiggen disagreed with Ms. Martin’s AOL projections and concerns, and that thisa supposed disagreement was based on reasonable differences in projections for the America Online business. (See id.; Rodon Decl. Ex. D (the Corrected Expert Report of Rene M. Stulz, dated July 8, 2008 (the “Stulz Corrected Rpt.”)) ¶ 28(b).) Professor Black notes that although AOL’s prospects deteriorated over the course of 2001, CSFB’s contemporaneous internal documents and Professor Black’s financial analyses suggest that both Ms. Martin and Mr. Kiggen did not believe CSFB’s public statements about AOL, but that they nonetheless failed to significantly revise their estimates or reduce CSFB’s $75 target price until September 25, 2001. (See Black Rpt. 22-23.) Professor Black disagrees with Professor Stulz’s justification for this target price, and he opines that a clear statement from CSFB to the effect that AOL’s performance would be substantially impacted by the advertising slowdown would have significantly affected investor views about AOL. (See id. at 24.)
Additionally, contrary to Defendants’ claim that Professor Black agreed with Professor Deighton regarding both of Professor Deighton’s opinions, (Defs.’ Mem. 40-41), Professor Black’s opinions also properly rebut Professor Deighton’s report. While Professor Black acknowledges that there was some general information about an advertising decline available to investors, he refutes Professor Deighton’s view that the market was already aware of the likely impact of the advertising decline on AOL. (See Black Rpt. 17-21; Deighton Rpt. ¶ 40.) Professor Black explains that the known information concerning an advertising decline was “soft” and uncertain early in the Class Period, particularly with regard to online advertising. (See Black Rpt. 18.) Moreover, Professor Black points out that in early 2001, AOL repeatedly  claimed that the advertising slowdown was not impacting AOL, and CSFB stood behind these statements despite their internally discussed concerns to the contrary. (See id. at 19 (noting a CSFB-hosted lunch with Robert Pittman, AOL’s chief operating officer, on May 22, 2001, and the subsequent, positive CSFB report on AOL).)
Professor Black also disputes Professor Deighton’s theory that despite the decline in the advertising markets, CSFB’s optimistic projections for AOL were reasonable. (See id.) More particularly, given that CSFB forecasted growth of the online advertising market overall at only 8%, Mr. Kiggen’s estimates that AOL would grow its own online ad revenue by 64% for the same time period were not plausible. (See id. at 20-21.) It was also implausible that in a maturing market AOL would capture 80% of the total worldwide growth in online advertising over the next five years, as CSFB projected. (Id. at 21.) Considering that advertising was the principal driver of growth and accounted for most of AOL’s profit, AOL’s subscription business, which was maturing and facing a competitive threat from broadband internet access, did not “justify anything close to a 60x EBITDA multiple.” (See id. at 19-20.) Thus, Professor Black concluded that Professor Deighton was wrong in opining that CSFB’s highly optimistic projections for AOL were reasonable.
In addition, Professor Black opines that since even Professor Deighton acknowledges that Mr. Kiggen lacked expertise regarding Time Warner, it was unreasonable for Mr. Kiggen to increase revenue projections for Time Warner businesses in the April 10, 2001 report in order to make up for a shortfall in the America Online business and maintain overall guidance. (See id. at 18.) Thus, even assuming Ms. Martin’s views were limited to Time Warner at the time, it was nonetheless unreasonable for CSFB’s published reports to be as optimistic as they were about AOL as a whole. (See id.)
 C. Professor Black’s Opinions Are The Product of Reliable Principles And Methods Which He Properly Applied To The Facts of This Case
Professor Black’s opinions are based primarily on his analysis of certain data contained in AOL’s financial releases compared to Defendants’ projections for AOL. (See Black Rpt. 7-16.) In particular, Professor Black focused on two detailed sets of CSFB projections, a Fidelity presentation on January 24, 2001 and a public report dated April 10, 2001. (Id. at 8.) Professor Black calculated Mr. Kiggen’s projected revenue growth rate for AOL’s fourth quarter of 2000, by subtracting Mr. Kiggen’s total revenue estimate for 2000 from the sum of AOL’s first three quarters, which were already known at the time of the January 24, 2001 presentation to Fidelity. Id. at 8-9. According to Professor Black’s calculations, Mr. Kiggen projected revenues of $2,208 for AOL’s fourth quarter of 2000. (Black Dep. Tr. 102:10-23.) By dividing Mr. Kiggen’s fourth quarter results by his projected third quarter results, Professor Black calculated Mr. Kiggen’s highly optimistic 11.8% quarter-over-quarter projected growth rate for AOL. (Black Dep. Tr. 102:16-23; 103:20-21.)
Moreover, Professor Black notes that by the time CSFB published its April 10, 2001 report, AOL had already failed to meet Mr. Kiggen’s revenue estimates for the fourth quarter of 2000. (Black Rpt. 11.) To account for this shortfall, Mr. Kiggen reduced his estimates for AOL revenue by $1.2 billion in this report; however, he increased Time Warner’s estimated revenue by $800 million at the same time, despite Ms. Martin’s insistence that Time Warner numbers should be brought down, in order to minimize the net reduction to AOL’s revenue overall to only $400 million. (Id.) Based on these analyses, Professor Black found that Defendants’ projections were overstated to a sufficient degree to suggest their falsity or bad faith, and that the overstatement was significant enough that it would likely have been important to investors.
Defendants attack Professor Black’s analysis, significantly exaggerating the effect and implications of a single, isolated error in one of Professor Black’s charts – namely, the accidental use of a restated 10-K instead of the original 10-K he had intended to use – on Professor Black’s results and the ultimate conclusions he drew from the chart. As his testimony and revised report show, Professor Black stood by his ultimate conclusions, he merely modified his language, changing his description of Mr. Kiggen’s projections from “absurd” to “highly optimistic.” (Black Dep. Tr. 104:2-10; 168:12:18.) As regrettable as any mistake is, this one error does not render his entire report unreliable. There is no reason to believe that any further inaccuracies exist in Professor Black’s data, considering that Defendants did not raise any at his deposition or in their brief.
At most, this error provides a basis for Defendants to argue to a jury that Professor Black’s opinions should be discounted, but it does not impact the admissibility of his testimony. See U.S. v. Vargas, 471 F.3d 255, 264 (1st Cir. 2006); see also Freeland v. Iridium World Comm’n, Ltd., 545 F. Supp. 2d 59, 88 (D.D.C. 2008) (“‘[T]he factual basis of an expert opinion goes to the credibility of the testimony, not the admissibility,’ making it a jury determination.”) (quoting Sappington v. Skyjack, Inc., 512 F.3d 440, 450 (8th Cir. 2008)). Indeed, for purposes of admissibility:
The issue is whether ‘(a) the opinions and conclusions of the expert are accompanied by information that enables the factfinder to evaluate the likely accuracy of the expert’s opinion, and (b) the information is presented in such a way that factfinders will not be fooled into excessively overvaluing the testimony’. . . Expert evidence should not be excluded merely because witnesses practicing in that field make errors with some frequency . . . but also because the factfinder has no information about the likelihood of error in the opinions, and thus cannot adjust the weight to be given to the evidence.
 U.S. v. Green, 405 F. Supp. 2d 104, 119 (D. Mass. 2005) (Gertner, J.) (internal citations omitted). Given the record of Professor Black’s deposition, Defendants can hardly claim that the jury might be “fooled in to excessively overvaluing” Professor Black’s testimony or that they will have “no information about the likelihood of error” in his opinions from which they may adjust the weight given his report. Hence, Defendants are free to “explore and expose any weaknesses” in the factual basis of Professor Black’s opinions during cross-examination, Int’l Adhesive Coating Co. v. Bolton Emerson Int’l, Inc., 851 F.2d 540, 545 (1st Cir. 1988), and their arguments to preclude his testimony based on this error should be rejected.
For all of the foregoing reasons, Defendants’ motion to preclude the expert testimony of Dr. Scott Hakala, Dr. M. Laurentius Marais, Professor Reinier Kraakman and Professor Bernard Black, should be denied in its entirety.
DATED: June 5, 2009 KAPLAN FOX & KILSHEIMER LLP
/s/ Frederic S. Fox
Frederic S. Fox (admitted pro hac vice)
Joel B. Strauss (admitted pro hac vice)
Donald R. Hall (admitted pro hac vice)
Melinda D. Rodon (admitted pro hac vice)
850 Third Avenue, 14th Floor
New York, NY 10022
Lead Counsel for Class Plaintiffs
nbsp;Edward F. Haber (BBO #215620) Todd Heyman (BBO #643804) SHAPIRO HABER & URMY LLP
53 State St. 37th Floor
Boston, MA 02109
Liaison Counsel for Class Plaintiffs
 Any challenge to Dr. Hakala’s impressive credentials would be futile. In 1989, Dr. Hakala received his Ph.D. in Economics from the University of Minnesota, where he wrote his dissertation under the direction of Edward Prescott, the recipient of the Nobel Prize in Economics in 2004. (Damages Rpt. Ex. A.) Dr. Hakala has published peer-reviewed articles on subjects including the economics of loss causation and the valuation of distressed equity securities, and has taught classes on asset pricing and market efficiency at the doctoral level. (Id.) Additionally, he is a director of CBIZ Valuation Group, LLC, one of the largest business valuation and consulting firms in the United States with offices in Dallas, Chicago, Atlanta, Milwaukee, St. Louis and Princeton, and has served as a consultant and expert witness on numerous occasions in litigations similar to this one. (Id.) Finally, having received the professional designation of Chartered Financial Analyst, Dr. Hakala is uniquely qualified to opine on numerous issues commonly confronted by equity research analysts.
 As explained more fully below, infra note 14, by Defendants’ own description of event study methodology, their study underestimates the impact of all news other than CSFB’s reports.
 (See, e.g., Damages Rebuttal Rpt. ¶¶ 41-42 & n.24); see also Nihat Aktas et al., Event studies with a contaminated estimation period, 13 J. Corp. Fin. 129 (2007) (highlighting “the importance of explicitly controlling for unrelated events occurring during the estimation window” of an event study”); John D. Jackson et al., The Impact of Non-Normality and Misspecification on Merger Event Studies, 13:2 Int. J. of the Econ. of Bus. 247, 262 (2006) (recommending controlling for all related and unrelated events because failing to do so causes “the true significance level of the test [to be] misstated”); see generally Philip Hans Franses, Time Series Models for Business and Economic Forecasting (Cambridge Univ. Press 2002) (1998).
 Professor Kraakman’s publications on the topics he opines on here have been cited in numerous court opinions, including the Supreme Court’s decision in Basic and the First Circuit’s opinion in PolyMedica. See Basic Inc. v. Levinson, 485 U.S. 224, 253 (1988); In re PolyMedica Corp. Sec. Litig., 432 F.3d 1, 9 (1st Cir. 2005).
 Rule 702 provides: “If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise, if (1) the testimony is based upon sufficient facts or data, (2) the testimony is the product of reliable principles and methods, and (3) the witness has applied the principles and methods reliably to the facts of the case.”
 Additionally, the Supreme Court articulated five non-exclusive factors that bear on the question of the admissibility of scientific or technical testimony: 1) whether the expert's technique or theory can be or has been tested; 2) whether the expert's technique or theory has been subject to peer review and publication; 3) the known or potential error rate; 4) the existence and maintenance of standards and controls; and 5) whether the technique has gained general acceptance in the relevant scientific community. Daubert, 509 at 593-94; Kumho Tire, 526 U.S. 137.
 In re Xcelera.com Sec. Litigation, Civ. Action No. 00-11649-RWZ, 2004 U.S. Dist. LEXIS 29064 (D. Mass. Sept. 30, 2004) (rejecting attacks on Dr. Hakala’s qualifications and his event study analysis and granting class certification); In re Xcelera.com Sec. Litig., 430 F.3d 503, 512-514 (1st Cir. 2005) (describing Dr. Hakala’s “sophisticated event study” and upholding the district court’s grant of class certification based on the study, in part.)
 (See Affidavit of Professor Matthew Richardson, In re Xcelera.com Sec. Litig., Civ. Action No. 00-11649- RWZ (D. Mass. filed March 7, 2003) (Dkt. Entry 73).)
 Defendants claim that just as in Xcelera, Dr. Hakala dummied out all company-specific news in his event studies here. (Defs.’ Mem. 21.) This claim is demonstrably false: not only has Dr. Hakala described at length his protocol for selecting only significant company-specific news, there was, in fact, AOL-specific news on every trading day of the study period here, but Dr. Hakala did not dummy out 168 days of his Damages Study. (See also Hakala 2008 Dep. Tr. 379:16-380:12 (testifying that there was AOL-specific news on every single day of the study period and that he was “much more selective here” in identifying significant news).) This procedure is perfectly in accord with the event study procedure Judge Zobel acknowledged as proper, observing that “the academic literature supports the use of dummy variables for events in which significant company-specific news is released[.]” In re Xcelera.com Sec. Litig., 2008 U.S. Dist. LEXIS 77807, at *3.
 Confronting the New Challenges of Scientific Evidence, 108 Harv. L. Rev. 1532, 1552 (May 1995) (defining “power” in the context of statistical significance tests).
 Moreover, all three of these studies identified as “events” all firm-specific news from the cited sources and controlled for all such news. Although Dr. Hakala did not control for all AOL-specific news here, these studies nonetheless evidence that there would have been no error had he done so.
 In fact, as explained more fully in the memorandum submitted in support of Plaintiffs’ Motion to Preclude the Expert Opinions of René M. Stulz and John Deighton, it is Professor Stulz’s event study method that is biased precisely because it fails to control for known company-specific news events, while Dr. Hakala eliminates this bias by properly controlling for all predetermined “material” events. (See Plts.’ Mot. to Preclude Expert Opinions 7-9.)
 Further, Dr. Marais testified that he had “no reason to expect” that there was any effect on the rate of “false positive” results due to Dr. Hakala’s use of dummy variables for event dates other than those dates under analysis. (Marais Dep. Tr. 167:2-168:4.)
 By Defendants’ own admission, Professor Stulz’s event study is not a valid measure of the statistical significance of any news other than CSFB’s reports because his study did not control for the possible effects of any other news. As Defendants explained, “it is important for the economist to exclude the possible effects of the events for which he or she is testing; otherwise, the predicted return will already include such effects, if any[.]” (Defs.’ Xcelera Mem. 4 (Dkt. entry 232) (emphasis added).) Hence, Professor Stulz’s event study systematically underreports the significance of all news other than CSFB’s AOL reports and cannot serve as a test of the statistical significance for such news.
 As Dr. Marais testified, the criterion for determining whether Dr. Hakala’s use of dummy variables comports with the relevant academic literature “would not be the mechanistic and superficial percent of some universe of observations that he might have been able to use” but would instead “go to valid technical principles concerning the validity of the exercise[.]” (Marais Dep. Tr. 105:5-106:13.)
 Dr. Hakala’s method of substituting dummy variables for days when significant news is released is the same as the method used by Aktas and referred to as the “manual” or “brute force” method. Aktas used the brute force method as a measure of the effectiveness of various other mathematical models for accomplishing the same thing, namely, eliminating the effects of the contaminating news events from their study. While the brute force method is considered highly effective, it may be impractical in the typical academic event study context because, as Aktas noted, the process of identifying all significant events may be “unreasonable for large-sample analyses” involving multiple companies and hundreds or thousands of observations. Aktas, supra note 3, at 130.
 See, e.g., Richard Roll, R2, 43 J. Fin. 541, 558 (1988) (“On average over [the sample of] ninety-six stocks, 23.7 percent of the daily observations were excluded by being either the day of a news event or the preceding day[,]” based on company-specific news listed in the Wall Street Journal and Dow Jones news service); Robert B. Thompson II et al., The Influence of Estimation Period News Events of Standardized Market Model Prediction Errors, 63 Acct. Rev. 448 (1988) (identifying and controlling for all company-specific news found in the Wall Street Journal index on 12.3% of trading days for a set of 2,358 selected stocks).
 See, e.g., John Y. Campbell et al., The Econometrics of Financial Markets, 524 (Princeton Univ. Press 1997) (discussing “the need for an a priori framework or specification for the model before confronting the data. By proposing such a specification . . . the chance of coming upon a spuriously successful model is reduced.”); see also Roll, supra note 17, at 558 (applying the same a priori approach); Thompson, supra note 17 (same).
 See Self-Regulatory Organizations; Notice of Filing of Proposed Rule Change by the National Association of Securities Dealers, Inc. Relating to Issuer Disclosure of Material Information, 67 Fed. Reg. 142 (Aug. 2, 2002), at 51306-10.
 Paul Ryan & Richard J. Taffler, Are Economically Significant Stock Returns and Trading Volumes Driven by Firm-specific News Releases?, 31(1)-(2) J. Bus. Fin. & Acct. 49, 74-76 (2004) (appendix providing a description of the news categories driving price and trading volume activity).
 See also Peter Kennedy, A Guide to Econometrics, 253-54 (MIT Press 5th ed. 2003) (“An F test would be used to test if several observations could be considered consistent with the estimated equation.”)
 Dr. Hakala likewise performed an F-test with respect to the events identified in his event study concerning market efficiency, and similarly found that the events identified in that study were significant at a 99.99% confidence level. (Market Efficiency Rpt. ¶ 17.)
 This argument is not only counterfactual relative to Dr. Hakala’s description of his methodology, but also ignores Dr. Hakala’s testimony that in preparing his Damages Study, he tested the variation in applying this protocol to the same set of source documents and found a very insignificant degree of difference in the implementations, around 2-3%. Dr. Hakala arrived at this conclusion through a double-blind test in which he and two of his staff members independently reviewed the gathered documents and identified conforming news, then Dr. Hakala compared the results of the separate reviews. (Hakala 2008 Dep. Tr. 190:23-192:9.) By contrast, Defendants’ “replication” arguments compare implementation of the document identification protocols to different sets of source documents for different purposes.
 Defendants complain that an additional 5 days when CSFB issued reports are found to be statistically significant in Dr. Hakala’s Damages Study versus his Market Efficiency Study, but they fail to note that there are also 3 days when CSFB issued reports that show decreased statistical significance findings in his Damages Study.
 (See Rodon Decl. Ex. I (the Affidavit of Scott D. Hakala filed in In re AOL Time Warner, Inc. Sec. & “ERISA” Litig., MDL No. 1500 (S.D.N.Y. Nov. 2, 2004) (the “Hakala AOL Aff.”)).)
 As Dr. Hakala explained in his Damages Rebuttal Report, “the original event study in the AOL Time Warner case was preliminary in nature, performed prior to discovery, and did not focus specifically on the impact of analyst reports on AOL’s share price.” (Hakala AOL Aff. ¶ 35 (emphasis added).)
 Similarly, the Damages event study in this litigation was of considerably broader scope than the Market Efficiency event study, providing evidence not only that AOL traded in an efficient market, but evidence relevant to the issues of materiality, loss causation and damages. As Dr. Hakala testified, this broader scope required more thorough searches for relevant disclosures to include. (Hakala 2008 Dep. Tr. 194:10-195:17.)
 See, e.g., Frank J. Fabozzi et al., Financial Modeling of the Equity Market: From CAPM to Cointegration, 431-32 (Wiley 2006) (“[A]ny process of model selection must start with strong economic intuition. . . . Economic intuition clearly entails an element of human creativity. As in any other scientific and technological endeavor, it is inherently dependant on individual abilities.”); John Y. Campbell et al., The Econometrics of Financial Markets, 524 (Princeton Univ. Press 1997) (stating that in identifying events, it is important “to impose some discipline on the specification search by a priori theoretical considerations[,]” which “may be in the form of well-articulated mathematical models of economic behavior, or behavioral models motivated by psychological phenomena, or simply heuristic rules of thumb based on the judgment, intuition, and past experience.”).
 (See also Damages Rebuttal Rpt. ¶ 35 (stating that Defendants’ criticism of the change in event window “ignores the fact that the [c]lass [p]eriod and loss causation analysis in the original AOL Time Warner securities litigation ranged from 1999 through August 2002.”).)
 Furthermore, both of Dr. Hakala’s market models are more powerful that Professor Stulz’s, which only explained 45.9% of the daily variance in AOL stock prices. (Stulz Decl. 9 n.14.)
 Dr. Hakala calculated damages according to the “out-of-pocket” rule, which “is well recognized and has been consistently applied by experts in securities litigation.” (Damages Rpt. 34 n. 24 (citing references).) See also John Finnerty & George Pushner, An Improved Two-Trader Model for Estimating Damages in Securities Fraud Class Actions, 8 Stan. J.L. Bus. & Fin. 213, 219 (2003) (describing the steps of a damages analysis under this method); Bradford Cornell & R. Gregory Morgan, Using Finance Theory to Measure Damages in Fraud on the Market Cases, 37 UCLA L. Rev. 883, 894 (1990) (discussing damages calculations models and jurisprudence); Michael Barclay & Frank C. Torchio, A Comparison of Trading Models Used for Calculating Aggregate Damages in Securities Litigation, 64(2)-(3) L. & Contemporary Probs. 105, 106 (2001) (“In general, damages per share are calculated as the artificial inflation when the shares were purchased minus the artificial inflation when the shares were sold.”)
 The so-called “analyst proxy” analysis, as described in greater detail, infra pp. 31-33. In an abundance of conservatism, rather than relying on any one analyst or third party disclosure concerning AOL’s underperformance as a basis for estimating the related fraudulent inflation, Dr. Hakala considered a set of 43 relevant analyst disclosures within the Class Period and applied peer-reviewed calibration techniques to prepare “an exact calculation of what portion of all those [relevant] events should be attributable to CSFB by itself.” (Hakala 2008
 Defendants apparently take issue with this long-standing, bedrock method of estimating damages, which they describe as “nonsensical” because it supposedly attributes accounting investigation-related inflation to AOL’s share price on January 12, 2001, long before Defendants learned of AOL’s accounting investigation on July 11, 2001. (Defs.’ Mem. 17 n.13.) This claim is simply false, as made clear from the Damages Report and accompanying exhibits: Dr. Hakala did not attribute any fraudulent inflation related to Defendants’ failure to disclose or adjust their estimates based on their knowledge of AOL’s accounting investigation until July 12, 2001, the day after Defendants actually gained knowledge of the accounting issues. (Damages Rpt. Exs. C-2 & C-2a.)
 As discussed in an article co-authored by Dr. Hakala, a variety of market, industry and other forces including “leakage” events or other relevant disclosures, act on the amount of fraudulent inflation in a stock on any given day in the class period. See Madge S. Thorsen, Richard A. Kaplan & Scott Hakala, Rediscovering the Economics of Loss Causation, 6 J. Bus. & Sec. L. 93 (2006) (discussing various ways that fraudulent inflation and dissipation occur, as well as how to model this inflation/dissipation with event studies); see also John Finnerty & George Pushner, An Improved Two-Trader Model for Estimating Damages in Securities Fraud Class Actions, 8 Stan. J.L. Bus. & Fin. 213, 219 (2003) (discussing adjusting the corrective events over time for a “comparable-stock index that recognizes both industry and market-wide influences” and adjusting for “firm-specific factors that can be directly attributed to company announcements that are not related to the fraud” using the “backwardization” approach based on percentage returns). Thus, the “relevant events” are not necessarily “corrective disclosures,” as Defendants appear to assume, but they may nevertheless effectively dissipate fraudulent inflation from the stock.
 See Henry J. Cassidy, Using Econometrics: a Beginners Guide 246-48 & 250-53 (1981) (stating, “to exclude some of the seasonal dummies because their estimated coefficients have low t-statistics is not recommended[,]” and discussing the related problems); Intriligator, Econometric Models, Techniques, and Applications 188-89 (1978) (“[E]xcluding relevant variables yields biased and inconsistent estimators, while including irrelevant variables yields unbiased and consistent estimators. Thus, in terms of bias and consistency, it is better to include too many than to include too few explanatory variables. . . . Considerable judgment, in fact, is called for in the specification of the model, balancing between including ‘too few’ and ‘too many’ variables[.]”); and Robert S. Pindyck & Daniel L. Rubinfeld, Econometric Models and Economic Forecasts, 162-66 (3d ed., McGraw- Hill Inc. 1991) (1976) (“If we are unsure of which explanatory variables ought to appear in a model, we face several trade-offs. The analysis shows that the cost of excluding a variable which should appear in the model is bias and inconsistency. The cost of adding one or more irrelevant variables is loss of efficiency. . . . In general, the choice of model form must be made in terms of the bias-efficiency trade-off, with the result dependent on the objective. If accurate forecasting is the goal, minimizing mean square error is one reasonable objective, since it accounts for both bias and efficiency.”).
 The only exceptions are: (1) the disclosure by AOL of an ad revenue shortfall with its earnings release on July 18, 2001; and (2) the subsequent analyst commentary on the earnings release on July 19, 2001. Each of those events was significant in the damages analysis.
 As Dr. Hakala described in his Damages Report, relevant disclosures are disclosures relating to AOL’s underperformance and advertising revenue weakness, and, after July 11, 2001, to AOL’s undisclosed layoffs and accounting improprieties. (Damages Rpt. ¶ 9.) Additionally, in Exhibits B-1, C-1a and C-2a, Dr. Hakala denoted the category of relevant event as ad-related, analyst-related, layoff-related or accounting-related, and provided a corresponding description of the relevant news. Thus, any failure on Defendants’ part to comprehend why particular dates were identified as relevant by Dr. Hakala does not signify any shortcoming in Dr. Hakala’s report or analysis.
 Defendants’ assertion that Dr. Hakala “ignored” CSFB’s AOL reports is incorrect. Dr. Hakala “specifically considered each CSFB report in the context of when it was issued and what it said (such as whether there was a material change in revenue or EBITDA forecasts not already factored in by the news, whether the opinions significantly altered CSFB’s prior opinions, and whether and to what extent there was confounding information).” (Damages Rebuttal Rpt. ¶ 16.)
 See cases cited infra note 46 and accompanying text.
 See also Lentell v. Merrill Lynch & Co., 396 F.3d 161, 173 (2d Cir. 2005) (ruling that loss causation may be proven through “materialization of the risk”); In re Enron Corp. Sec., Derivative and ERISA Litig., No. MDL-1446, 2005 WL 3504860, at *16 (S.D. Tex. Dec. 22, 2005) (“[B]esides a formal corrective disclosure by a defendant . . . the market may learn of possible fraud from a number of sources [such as] whistleblowers, analysts’ questioning financial results, resignations of CFOs or auditors, announcements by the company of changes in accounting treatment going forward, newspapers and journals, etc.”).
 In support of their argument that Dr. Hakala’s treatment of confounding events renders his analysis unreliable, Defendants cite to the Omnicom court’s ultimate grant of summary judgment based, in part, on the fact that Dr. Hakala’s analysis in that case did not disaggregate any impact attributable to the “negative tone” of certain articles from the impact of the partial revelations of the truth contained in those same articles. In re Omnicom Group, Inc. Sec. Litig., 541 F. Supp. 2d 546, 553 (S.D.N.Y 2008). That decision is currently on appeal to the Second Circuit but in any event, is distinct from the present matter because unlike in Omnicom, here Dr. Hakala has disaggregated the impact of potentially confounding events by applying weights to reflect the impact of multiple news items and/or CSFB’s impact as distinct from that of all other speakers. See id. at 553 (“While plaintiffs need not quantify the fraud-related loss, they must ‘ascribe some rough proportion of the whole loss to [the alleged] misstatements.’”) Hence, Dr. Hakala’s anlysis here satisfies the standard articulated in Omnicom.
 As the Ninth Circuit has explained, “as long as the misrepresentation is one substantial cause of the investment’s decline in value, other contributing forces will not bar recovery under the loss causation requirement but will play a role in determining recoverable damages.” In re Daou Sys., Inc., 411 F.3d at 1025; see also Robbins v. Koger Props., Inc., 116 F.3d 1441, 1447 (11th Cir. 1997) (“Because market responses, such as stock downturns, are often the result of many different, complex, and often unknowable factors, the plaintiff need not show that the defendant's act was the sole and exclusive cause of the injury he has suffered; he need only show that it was ‘substantial’, i.e., a significant contributing cause.”)
 Dr. Hakala determined whether analyst reports were positive, negative or neutral based on the objective criteria of whether they contained material changes in the price targets, buy/sell recommendations or earnings forecasts. (Hakala 2008 Dep. Tr. 202:18-204:3.)
 Defendants suggest that Dr. Hakala’s determinations of whether particular events were confounded by multiple pieces of news or analyst reports are too subjective, pointing to his inclusion in his damages analysis of October 17, 2001, which he testified was “partially confounded,” as a supposed example of this flaw. (Defs.’ Mem. 10.) On October 17, 2001, AOL released third quarter 2001 earnings, and Merrill Lynch downgraded AOL from “buy” to “neutral” based on AOL’s disappointing ad revenue performance. Dr. Hakala explained that he included the date as a relevant analyst disclosure because the “news reports clearly attributed the decline in AOL’s share price to the Merrill Lynch report and not the earlier earnings release, (which was viewed as expected and neutral in its impact)[.]” Thus, Dr. Hakala’s inclusion of October 17, 2001 was clearly not based on any supposedly “subjective” beliefs about confounding events, but rather, on what the press reported as being the more significant item.
Moreover, as with other such events, Dr. Hakala only included a small proportion, 13.41%, of the total impact of the day’s news in his damages analysis.
 See also In re Motorola Sec. Litig., 505 F. Supp. 2d 501, 544 (N.D. Ill. 2007) (rejecting defendants’ arguments that the alleged curative disclosures were too vague because “[d]efendants’ proposed rule would provide an expedient mechanism for wrongdoers to avoid securities fraud liability.”); Freeland v. Iridium World Commc’n, Ltd., 233 F.R.D. 40, 47 (D.D.C. 2006) (following precedents rejecting an interpretation of Dura which “would allow wrongdoers to immunize themselves with a protracted series of partial disclosures.”); In re Vivendi Universal, S.A. Sec. Litig., No. 02 Civ. 5571 (RJH), 2009 U.S. Dist. LEXIS 34563, *43 (S.D.N.Y. Apr. 6, 2009) (observing that loss causation “determinations may often rest in part on legal policy considerations[.]”) (internal citations omitted).
 See also Vivendi, 2009 U.S. Dist. LEXIS 34563, at *43 (stating that “stock prices decline in reaction to information released into the market rather than in reaction to the fraudulent statements themselves. When that information was previously concealed from the market by the fraud, we can properly conclude that whatever decline in the stock price was caused by the release of that information was also caused by the fraud.”); Nathenson v. Zonagen, 267 F.3d 400, 419 (5th Cir. 2001) (“We also realize that in certain special circumstances public statements falsely stating information which is important to the value of a company’s stock traded on an efficient market may affect the price of the stock even though the stock’s market price does not soon thereafter change. For example, if the market believes the company will earn $1.00 per share and this belief is reflected in the share price, then the share price may well not change when the company reports that it has indeed earned $1.00 a share even though the report is false in that the company has actually lost money (presumably when that loss is disclosed the share price will fall”). To the extent Defendants’ argument rests on the premise that a different legal standard should be applied to analysts, that position has been discredited. In re Salomon Analyst Metromedia Litig., No. 06-3225-cv, 2008 WL 4426412, at *7 & *9 (2d Cir. Sept. 30, 2008) (ruling that “no heightened test is needed in the case of research analysts” for application of the fraud-on-the-market presumption).
 Indeed, in attacking Dr. Hakala’s analyst proxy analysis, Professor Stulz incredibly suggests that a more accurate damages estimate could be derived using the impact of ten “negative” AOL reports issued by Defendants
 The analyst proxy analysis does not apply to the accounting or layoff claims, for which damages were calculated based solely on the curative disclosures identified in the complaint. (See Damages Rpt. ¶¶ 10-11.)
 Dr. Hakala confirmed his identification of significant analyst statements using a joint t-test, which demonstrated that the identified statements were extremely significant, with a joint t-statistic of 12.28 for all statements and 9.12 for negative analyst statements. (Damages Rpt. ¶¶ 27-28.)
 As discussed supra pp. 26-27, two substantially similar analyst reports were issued by highly respected analysts on September 19, 2001: one by Bear Stearns and the other by CSFB. Thus, Dr. Hakala divided the AOL impact of the positive analyst commentary for that day evenly between the two firms.
 Professor Stulz’s regression analysis measured a significantly higher one-day impact on February 5, 2001 of 3.37%, but a somewhat lower one-day impact on September 19, 2001, of which 1.55% is attributable to CSFB, yielding a slightly higher average CSFB inflationary impact of 2.46%, relative to the 2.35% Dr. Hakala found. (See Rodon Decl. Ex. L, (attaching Exhibit 2 from the June 21, 2007 deposition of Rene M. Stulz).)
 Defendants misrepresent this aspect of the DeMarco ruling, suggesting that the DeMarco court was critical of the sample size of the study of 200 companies, when in fact, the sample size criticism applied to the estimation of maximum inflation, which was limited to three events. DeMarco, 222 F.R.D. at 249; (Defs.’ Mem. 30). Moreover, Defendants erroneously state that the study referenced in DeMarco was of 200 companies over three years, when in fact, it was over two years (1989-1991). DeMarco, 222 F.R.D. at 248.
 Published in Litigation Services Handbook: The Role of the Financial Expert, 2005 Cumulative Supplement, 3d ed., John Wiley & Sons.
 Further, Dr. Marais could hardly be as “expert” in the factual application of Dr. Hakala’s event study as Dr. Hakala is himself. Instead of improperly offering “me too” testimony to bolster the various professional judgments by Dr. Hakala that Defendants’ challenge, Dr. Marais guides the jury concerning the key principles in making their own evaluation of Dr. Hakala’s event study. For example, Defendants criticize Dr. Marais for not opining on whether one step the event study is “replicable” based on a purported percentage of variance in implementation, however, Dr. Marais testified that no artificial percentage of variance can answer the question because the relevant issue is whether that variance translates into a different ultimate outcome for the event study as a whole. (Marais Dep. Tr. 87:2-89:23.) Thus, Dr. Marais appropriately provides the jury with guidance concerning the technically important factors for evaluating Dr. Hakala’s event study, rather than instructing the jury on what its opinion should be, as Defendants implicitly suggest he should have done.
 Although Professor Kraakman testified that he is “[p]robably not” more of an expert than defense counsel on the academic literature pertaining to analysts, he did qualify the statement, adding, “I don't know what your background is and how used you are to reading finance articles[.]” (Kraakman Dep. Tr. 52:5-7.)
 Formally, Professor Kraakman’s report examines four key issues: (i) whether the reports of prominent securities analysts can have a material effect on share prices in an efficient market; (ii) whether, assuming such an effect, there is justification for treating analyst statements differently than issuer statements for purposes of fraud- on-the-market theory; (iii) whether prices in an efficient market react only to new information bearing on share value; and, if so, (iv) whether prices in an efficient market necessarily react to statements by analysts who merely repeat information that is already fully disclosed to the market. (Kraakman Rpt. ¶ 3.)
 In Nieves-Villanueva, cited by Defendants, the expert at issue testified as to “the holdings of various opinions of the Supreme Court of Puerto Rico” and drew legal conclusions that the First Circuit described as “misleading at best.” Nieves-Villanueva v. Soto-Rivera, 133 F.3d 92, 99 (1st Cir. 1997).
 The remaining cases cited by Defendants in support of their argument that Professor Kraakman opines on issues of law are also inapposite. See Pelletier v. Main Street Textiles, LP, 470 F.3d 48, 55 (1st Cir. 2006) (Expert testimony interpreting the meaning of OSHA regulations excluded.); U.S. v. Buchanan, 964 F. Supp. 533, 537 (D. Mass. 1997) (Expert testimony excluded to the extent that it “recounts . . . the standards of conduct that plainly derive directly from the statutes and regulations[.]”); Interfaith Cmty. Org. v. Honeywell Int’l, Inc., 399 F.3d 248, 253-54 (3rd Cir. 2005) (The majority held that determinations under the Resource Conservation Recovery Act are a question of fact, but Defendants cited the concurrence, which argues that these determinations should be reviewed as a mixed question of fact and law. Id. at 268-69.).
 See, e.g., Information Asymmetry, the Internet, and Securities Offerings, 2 J. Small & Emerging Bus. L. 91-99 (1998) (http://ssrn.com/abstract=84489).
 As Professor Black testified, he had intended to use the original financials because he was attempting to show that Defendants’ projections were highly optimistic based on the information they had at the time. (See Black Dep. Tr. 93:4-9, 97:4-7.)