Efficient semiparametric estimation in time-varying regression models

TL;DR

This paper develops efficient semiparametric estimation methods for linear regression models with time-varying coefficients, dependent regressors, and errors, improving estimation efficiency under certain error structures.

Contribution

It introduces a new estimation procedure for non time-varying parameters at root n rate, surpassing previous methods in efficiency for specific error models.

Findings

Efficient estimator for models with martingale difference errors.

Asymptotic efficiency achieved for Gaussian error models.

Derived an efficient information matrix for time-varying AR processes.

Abstract

We study semiparametric inference in some linear regression models with time-varying coefficients, dependent regressors and dependent errors. This problem, which has been considered recently by Zhang and Wu (2012) under the functional dependence measure, is interesting for parsimony reasons or for testing stability of some coefficients in a linear regression model. In this paper, we propose a different procedure for estimating non time-varying parameters at the rate root n, in the spirit of the method introduced by Robinson (1988) for partially linear models. When the errors in the model are martingale differences, this approach can lead to more effcient estimates than the method considered in Zhang and Wu (2012). For a time-varying AR process with exogenous covariates and conditionally Gaussian errors, we derive a notion of efficient information matrix from a convolution theorem adapted to triangular arrays. For independent but non identically distributed Gaussian errors, we construct an asymptotically efficient estimator in a semiparametric sense.