This paper extends a recent study by Malatesta [14] on measuring abnormal performance using joint generalized least squares. For monthly data and a random sample of securities, Malatesta finds that there is little benefit in using more sophisticated econometric techniques to identify abnormal returns. The current study extends these results using a design that is more amenable to the benefits of the generalized methods and is consistent with actual event studies. Most notably, the study uses a sample of securities experiencing an actual event and tests both monthly and daily data. In addition, iterative techniques are compared to the ordinary least squares and estimated generalized least squares methods. The results of this study support the original conclusions of Malatesta, indicating no measurable gain in using any of the systems methods for event study applications.

The traditional Capital Asset Pricing Model (CAPM) provides a foundation for the estimation of systematic risk that has been applied extensively in studies of investment performance, market efficiency, predictive models, and capital budgeting, to name only a few. Lee [10] considered a special case of nonlinearities occurring in the estimation of systematic risk within the context of the investment horizon problem. His findings, based on a limited sample, provided significant methodological implications for the estimation process and have received wide readership through republication of the study in a readings text [6].

This paper examines estimation issues associated with multivariate tests of asset pricing. Two issues are considered: (1) the constraint that the sample size (N) must be less than the time series (T), and (2) the relative effect on power of using the multivariate statistic versus a univariate counterpart. We find that an alternative statistic that allows for large N does not dominate the usual portfolio tests. More notably, we find that the power of a simple diagonal statistic usually dominates the multivariate statistic for cases considered in this study.

In business research, firm size is both ubiquitous and readily measured. Complexity, another firm-related construct, is also relevant, but difficult to measure and not well-defined. As a result, complexity is less frequently incorporated in empirical designs. We argue that most extant measures of complexity are one-dimensional, have limited availability, and/or are frequently misspecified. Using both machine learning and an application-specific lexicon, we develop a text solution that uses widely available data and provides an omnibus measure of complexity. Our proposed measure, used in tandem with 10-K file size, provides a useful proxy that dominates traditional measures.

Measuring the extent to which a firm is financially constrained is critical in assessing capital structure. Extant measures of financial constraints focus on macro firm characteristics such as age and size, variables highly correlated with other firm attributes. We parse 10-K disclosures filed with the U.S. Securities and Exchange Commission (SEC) using a unique lexicon based on constraining words. We find that the frequency of constraining words exhibits very low correlation with traditional measures of financial constraints and predicts subsequent liquidity events, such as dividend omissions or increases, equity recycling, and underfunded pensions, better than widely used financial constraint indexes.