inference in dynamic panel models accepted - www.pdfVIP

Statistical inference in dynamic panel data models Tze Leung Lai ∗,1 Department of Statistics, Stanford University, Stanford, CA, 94305-4065, USA. Dylan S. Small Department of Statistics, The Wharton School, University of Pennsylvania, Philadelphia, PA, 19104, USA. Jia Liu Department of Management Science and Engineering, Stanford University, Stanford, CA, 94305-4026, USA. Abstract T.W. Anderson and his collaborators have made seminal contributions to inference with instrumental variables and to dynamic panel data models. We review these contributions and the extensive economic and statistical literature that these con- tributions spawned. We describe our recent work in these two areas, presenting new approaches to (a) making valid inferences in the presence of weak instruments and (b) instrument and model selection for dynamic panel data models. Both approaches use empirical likelihood and resampling. For inference in the presence of weak in- struments, our approach uses model averaging to achieve asymptotic efficiency with strong instruments but maintain valid inferences with weak instruments. For instru- ment and model selection, our approach aims at choosing valid instruments that are are strong enough to be useful. Key words: Instrumental variables; Weak instruments; Model seelection; Empirical likelihood; Resampling. 1. Introduction and background A panel (or longitudinal) data set follows the same sample of individuals over time, and thus provides multiple observations on each individual in the sample. A ∗ Corresponding author. Email address: lait@stanford.edu (Tze Leung Lai). 1 Supported by the National Science Foundation Preprint submitted to Elsevier