Description
Book SynopsisEconometric Model Specification reviews and extends the author's papers on consistent model specification testing and semi-nonparametric modeling and inference. This book consists of two parts. The first part discusses consistent tests of functional form of regression and conditional distribution models, including a consistent test of the martingale difference hypothesis for time series regression errors. In the second part, semi-nonparametric modeling and inference for duration and auction models are considered, as well as a general theory of the consistency and asymptotic normality of semi-nonparametric sieve maximum likelihood estimators. Moreover, this volume also contains addendums and appendices that provide detailed proofs and extensions of all the results. It is uniquely self-contained and is a useful source for students and researchers interested in model specification issues.
Table of ContentsIntroduction; Consistent Model Specification Tests: Consistent Tests of Functional Form of Regression Models: the I.i.d. Case; Consistent Tests of Functional Form of Regression Models: the Time Series Case; A Consistent Conditional Moment Test of Functional Form; Asymptotic Theory of Integrated Conditional Moment Tests; Asymptotic Theory of Integrated Conditional Moment Tests; Semi-Nonparametric Modeling and Inference: Semi-Nonparametric Interval-Censored Mixed Proportional Hazard Models; Semi-Nonparametric Competing Risks Analysis of Recidivism; Semi-Nonparametric Auction Models; Consistency and Asymptotic Normality of Sieve ML Estimators Under Low-Level Conditions;
