Guides · Securities

Daubert Challenges to Event Studies in Securities Cases

The event study is the workhorse of securities litigation. Plaintiffs use it to show market efficiency, price impact, loss causation and damages; defendants use it to show the absence of each. Because so much rides on one regression, it is also a frequent target of Rule 702 motions on both sides. This guide covers the choices that draw challenges, how courts have ruled on them, and what the December 2023 amendment to Rule 702 changes about the argument.

Guide 1 of 4 · September 8, 2026 · 8 min read
On this page
  1. The standard
  2. Event selection
  3. Confounding news
  4. Windows, thresholds
  5. What to prepare

The gatekeeping standard after December 1, 2023

In Daubert v. Merrell Dow Pharmaceuticals, Inc., the Supreme Court held that "the trial judge must ensure that any and all scientific testimony or evidence admitted is not only relevant, but reliable." Kumho Tire Co. v. Carmichael held that this gatekeeping obligation "applies to all expert testimony," which brings financial economists squarely within it.

Rule 702 was amended effective December 1, 2023. As amended, it admits expert testimony only if "the proponent demonstrates to the court that it is more likely than not that" the testimony will help the trier of fact, rests on sufficient facts or data, is the product of reliable principles and methods, and reflects a reliable application of those methods to the facts. The Sixth Circuit, in Davis v. Sig Sauer, Inc., described the amendment as clarifying that the preponderance standard applies to Rule 702. The en banc Federal Circuit, in EcoFactor, Inc. v. Google LLC, a damages-expert case, said the amendment was meant "to clarify that the proponent of expert testimony bears the burden of establishing its admissibility and to emphasize that an expert's opinion must stay within the bounds of a reliable application of the expert's basis and methodology." It also quoted the Advisory Committee's view that rulings treating "the sufficiency of an expert's basis, and the application of the expert's methodology" as "questions of weight and not admissibility" are "an incorrect application of Rules 702 and 104(a)."

That point matters for event studies. The method itself is accepted. In Halliburton Co. v. Erica P. John Fund, Inc. (Halliburton II), the Supreme Court described event studies as "regression analyses that seek to show that the market price of the defendant's stock tends to respond to pertinent publicly reported events." The live disputes concern application: which dates, which benchmark, which window, which threshold. Before the amendment, some courts sent those disputes to the jury as matters of weight. The amended rule and its note put them in front of the judge.

The same standard applies at class certification. In Georgia Firefighters' Pension Fund v. Anadarko Petroleum Corp., the Fifth Circuit stated that "Daubert therefore applies with the same rigor at the class certification stage as at trial," and vacated a certification order because the district court had not fully considered the defendants' challenge to a rebuttal event study.

Event selection and the estimation window

An event study begins with the choice of event dates, and challenges often start there. In Bricklayers & Trowel Trades International Pension Fund v. Credit Suisse Securities (USA) LLC, the First Circuit affirmed exclusion of the plaintiffs' loss causation expert. The court observed that "[e]vent selection should not be difficult to understand," yet the study included dates unrelated to the complaint and treated some events as corrective that the complaint called inflationary. From all appearances, the court said, "the event study is more concerned simply with identifying abnormal market movement than in supporting the shareholders' causation allegations." The event list must track the pleaded theory, and the expert should be able to explain how each date was chosen without reference to the returns.

Bricklayers also addressed stale information. Several event dates rested on commentary about facts already public, and the court held that those items "did no more than to provide gloss on public information," so an efficient market would not have moved on them. A party that relies on market efficiency to certify a class cannot set it aside when proving loss causation.

The estimation window is the period used to fit the expected-return model, and the attack usually concerns what was removed from it. In Bricklayers, the expert had used dummy variables to exclude 211 of 388 days in the study period, and the defense argued that this deflated baseline volatility and made event-day returns look significant. The First Circuit was more receptive to the expert than the district court had been. It noted academic support for conditioning on news days and that the defendant's own study excluded dates for the same reason. It still found that the number of excluded days and the method for selecting them "create close questions," and it rested its holding on other grounds. Under the amended rule, a court may be less willing to treat an opaque exclusion rule as a question of weight. Defense counsel should ask for the written criteria used to exclude days and test whether another economist could reproduce the list. Plaintiffs' counsel should make sure those criteria exist before the report is served.

Confounding news

The most common substantive attack is that something other than the alleged corrective disclosure moved the price. In In re Williams Securities Litigation, the Tenth Circuit affirmed exclusion of the plaintiffs' loss causation expert because his theories could not separate fraud-related losses from losses caused by other news. The court stated the burden plainly: "The plaintiff bears the burden of showing that his losses were attributable to the revelation of the fraud and not the myriad other factors that affect a company's stock price." Bricklayers turned that burden into a methodological requirement: "when conducting an event study, an expert must address confounding information that entered the market on the event date." The expert there failed because he "seemingly made a judgment call as to confounding information without any methodological underpinning."

Confounding disputes increasingly turn on timing within the trading day. In Anadarko, the alleged corrective disclosure came out after the close, and news of an unrelated event followed shortly afterward. The plaintiffs' rebuttal report relied on after-hours price movement between the two releases and on a new event study controlling for the second item. The defendants argued that the expert had not shown the after-hours market was efficient, had not tested the after-hours move for statistical significance, had misidentified when the competing news became public, and had dropped the industry index from the revised study. The Fifth Circuit did not resolve those points, but it held that the district court had to give them a full Daubert analysis on remand.

For defense experts, the confounding argument has a limit at class certification. In Waggoner v. Barclays PLC, the defense economist attributed part of the price drop to concern about the regulatory action itself rather than to revelation of the truth. The Second Circuit held that "merely suggesting that another factor also contributed to an impact on a security's price does not establish that the fraudulent conduct complained of did not also impact the price of the security." Partial attribution may matter for damages. It does not rebut price impact.

Event windows, significance thresholds and the benchmark

The remand decision in Erica P. John Fund, Inc. v. Halliburton Co., from the Northern District of Texas, is among the most detailed judicial treatments of these choices.

Event windows. The plaintiffs' expert used two-day windows for several dates. The court held that "[a]bsent a compelling explanation, which was not given, the Court finds that the use of a two-day window is inconsistent with an efficient market, especially where the relevant disclosure was made before the market opened on Day 1." Multiple-day windows are not forbidden, but the expert must justify one with the timing of the release or evidence of delayed dissemination. A window that appears to have been chosen because the one-day return was not significant invites exclusion.

Significance thresholds. The court treated the 95 percent confidence level as necessary and held that an intraday reaction significant only at 90 percent did not establish price impact. It also accepted the defense economist's adjustment for multiple comparisons, concluding that "applying a Holm-Bonferroni multiple comparison adjustment is appropriate in this case," even though, as the court acknowledged, such adjustments are rarely used in securities event studies. Because the adjustment raises the bar for each individual date, plaintiffs' experts should be ready to address it directly. They should also be ready for the court's observation that some dates failed even without it.

The benchmark. The court adopted a peer index, built by the plaintiffs' expert from companies that analysts identified as peers, in measuring significance, because adding it increased the explanatory power of the defense model. A market model that tracks the broad market but not the company's industry, or that drops an industry index when other controls are added, as the defendants alleged in Anadarko, is an easy target.

Partial exclusion. Courts do not always exclude the whole study. In In re Pfizer Inc. Securities Litigation, after the district court held on summary judgment that two corrective disclosures could not be attributed to the defendant, the plaintiffs' expert proportionally reduced the price increases on other dates. The Second Circuit held that the district court could find that adjustment unreliable but abused its discretion by excluding all of the expert's testimony, because the flaw did not undermine the underlying event study. A motion aimed at one step often succeeds where an attack on the whole method would fail.

What to prepare

  • The event list, with the documented basis for each date and its link to the complaint.
  • Release timestamps for every disclosure and every potentially confounding item, including after-hours news.
  • Written criteria for any days excluded from the estimation window, and results with and without the exclusions.
  • The benchmark specification, including industry or peer indices and their explanatory power.
  • One-day and multiple-day results side by side, with the reason for any longer window.
  • Significance at the 95 percent level, with and without any multiple-comparison adjustment.
  • A list of the disputed steps that are application of method, which amended Rule 702 places with the court.

Supreme Mind AI builds Expert Witness Simulation. A lawyer points it at a matter and picks a class of expert, such as a defense event-study econometrician, and one run returns a brief in four sections: the expert's likely opinion, ranked cross-examination weaknesses, the Daubert and Rule 702 attack surface, and a directional settlement range. Experts are de-identified composites of a class, never a named individual, supporting quotations are checked verbatim against the public record, and nothing it produces is filed or testifies. A complete sample brief on In re Luckin Coffee is at /sample-brief, and the method is described at /method.

This guide is general information, not legal advice.

Sources

How each court of appeals applies amended Rule 702: see the Rule 702 tracker.

Run this analysis against the expert on your own matter.Start freeRead a sample brief →

Experts in this guide's practice

Defense event-study econometricianPlaintiff event-study and loss-causation econometricianMarket-efficiency expertSecurities accounting and GAAP expert

The series

01Securities · September 8, 2026Daubert Challenges to Event Studies in Securities Cases02Securities · September 15, 2026Price Impact at Class Certification After Goldman03Antitrust · September 22, 2026Common Impact and Pass-Through at Class Certification04Any expert · September 29, 2026Amended Rule 702 After Nearly Three Years: What Courts Exclude

Related: the expert class library, a complete sample brief, and how it works.