Partial linear models with general distortion measurement errors (Q2283581)

The paper under review discusses partial linear models that contain both multiplicative and additive distortion measurement errors. The author proposes a new calibration procedure for the unobserved variables (by coupling the conditional means and conditional variances). Then, a profile least-squares estimator and the confidence intervals of the parameter are recommended, and the related asymptotic results are derived. Next, for hypothesis testing of the parameters, a restricted estimator under the null hypothesis and a test statistic are proposed and their asymptotic properties are established. Further, the smoothly clipped absolute deviation penalty (to select relevant variables) is utilized. The resulting penalized estimators are shown to be asymptotically normal and have the oracle property. Later on, estimation, hypothesis testing, and variable selection are discussed under the scenario of multiplicative distortion alone. Finally, the illustration of proposed results is provided.