A0445
Title: Missing data in asset pricing panels
Authors: Michael Weber - Chicago Booth (United States) [presenting]
Andreas Neuhierl - Washington University in St. Louis (United States)
Joachim Freyberger - University of Bonn (Germany)
Bjoern Hoeppner - University of Bonn (Germany)
Abstract: Missing data for return predictors is a common problem in cross-sectional asset pricing. Most papers do not explicitly discuss how they deal with missing data but conventional treatments focus on the subset of firms with no missing data for any predictor or impute the unconditional mean. Both methods have undesirable properties - they are either inefficient or lead to biased estimators and incorrect inference. A simple and computationally attractive alternative is proposed using conditional mean imputations and weighted least squares, cast in a generalized method of moments (GMM) framework. This method allows the use of all observations with observed returns, it results in valid inference, and it can be applied in non-linear and high-dimensional settings. In Monte Carlo simulations, it is found that it performs almost as well as the efficient but computationally costly GMM estimator in many cases. The procedure is applied to a large panel of return predictors and finds that it leads to improved out-of-sample predictability.
