Calibration of the empirical likelihood for semiparametric varying-coefficient partially linear models with diverging number of parameters

Authors : Junwei HE, Xia CHEN

Pages : 230-241

View : 16 | Download : 6

Publication Date : 2019-02-01

Article Type : Research Paper

Abstract :This article is concerned with the calibration of the empirical likelihood for semiparametric varying-coefficient partially linear models with diverging number of parameters. However, there is always substantial lack-of-fit, when the empirical likelihood ratio is calibrated by a bias-corrected empirical likelihood, producing tests with type I errors much larger than nominal levels. So we consider an effective calibration method and study the asymptotic behavior of this bias-corrected empirical likelihood ratio function. Some simulation studies are conducted to illustrate our approach.
Keywords : Varying coefficient partially linear models, Empirical likelihood, Bias correction, Asymptotic normality, Coverage accuracy