Empirical likelihood for semiparametric varying-coefficient partially linear errors-in-variables models

In this paper, we apply the empirical likelihood method to semiparametric varying-coefficient partially linear errors-in-variables models. Under the two different identifying conditions, the two different empirical log-likelihood ratios for the parametric components were proposed and their asymptotic distributions were derived. By this, the two different empirical likelihood confidence regions for the parameter components with asymptotically correct coverage probabilities can be constructed. Simulations show that the proposed empirical likelihood confidence regions have satisfactory coverage.