CFDP 968
Tests of Specification for Parametric and Semiparametric Models
Author(s):Publication Date: January 1991
Pages: 80
Abstract:
This paper provides a general framework for constructing specification tests for parametric and semiparametric models. The paper develops new specification tests using the general framework. In particular, specification tests for semiparametric partially linear regression, sample selection, and censored regression models are introduced. The results apply in time series and cross-sectional contexts. The method of proof exploits results concerning the stochastic equicontinuity or weak convergence of normalized sums of stochastic processes.
Keywords:
Infinite dimensional nuisance parameter, semiparametric model, specification test, stochastic equicontinuity
JEL Classification Codes: 211
See CFP: 844