Statistical inference for a heteroscedastic regression model with φ-mixing errors

Abstract

In this paper, we mainly study the asymptotic normality of wavelet estimators in the heteroscedastic regression model with φ-mixing errors. Under some suitable conditions, the asymptotic normality of the wavelet estimators of g and f in the heteroscedastic regression model with φ-mixing errors are obtained, which generalize or improve the corresponding ones for α-mixing random variables and martingale difference sequence in some sense. Also, the simulation study of the finite samples provided in this paper shows the validity of our results.