EI Vice President Kevin Caves and Principal Hal Singer have published “Applied Econometrics: When Can an Omitted Variable Invalidate a Regression?” in the December 2017 issue of the ABA’s Antitrust Source. In the article, Caves and Singer explore “omitted variable bias,” a fundamental econometric concept that frequently arises when regression models are applied to assess liability and damages in antitrust litigation. The authors explain that, although every regression has some omitted variables, the relevant question is whether the omission generates bias that significantly compromises the reliability of the regression model. Using real-world examples, the authors examine the conditions under which this is the case. Because the direction of the bias is often highly relevant (e.g., are damages estimated conservatively, or are they over-stated?), Caves and Singer also explain how the direction of the bias can be predicted—and, in some cases, known with certainty.