Testing structural change in partially linear single-index models with error-prone linear covariates

Motivated by an analysis of a real data set from Duchenne Muscular Dystrophy (Andrews and Herzberg, 1985), we propose a new test of structural change for a class of partially linear single-index models with error-prone linear covariates. Based on the local linear estimation for the unknowns in these semiparametric models, we develop a new generalized F-test statistics for the nonparametric part in the partially linear single-index models with error-prone linear covariates. Asymptotic properties of the newly proposed test statistics are proved to follow asymptotically the chi-squared distribution. The new Wilks' phenomenon is unveiled in a class of semiparametric measure error models. Simulations are conducted to examine the performance of our proposed method. The simulation results are consistent with our theoretical findings. Real data examples are used to illustrate the proposed methodology.