I. Introduction

In a 1994 monograph, two of the authors criticized the use of trading models to estimate aggregate damages in open-market class action securities fraud litigation.[1] We argued that trading models often would not provide reliable estimates of aggregate damages, and noted that the then-available empirical evidence suggested that such models would typically overstate damages, sometimes by as much as 74%.[2] In light of these conclusions, we argued that pretrial discovery of actual trading data would provide litigants and courts with much more reliable information, and that damage awards at trial should be based on actual trading data.3

This monograph assesses the developments that have occurred in the intervening years. In Section II, we examine recent court decisions that have precluded the admission of evidence concerning aggregate damages based on trading models on Daubert grounds, and also found that an adequate remedy could be fashioned by having the jury determine a per-share damage amount per day, and then using the claims administration process to calculate an accurate, reliable total damage figure and payout.[4] In Section III, we review and critique articles published after 1994 that advocate the use of trading models. In Section IV, we present some powerful new evidence demonstrating the unreliability of trading models as estimators of aggregate damage claims. Finally, in Section V, we discuss some of the implications of our conclusions.

 Click here to view the full text of this Briefly as an Adobe Acrobat PDF.