A0770
Title: Extrapolation estimation for nonparametric regression with measurement error
Authors: Weixing Song - Kansas State University (United States) [presenting]
Abstract: The purpose is to introduce an extrapolation algorithm designed for estimating regression functions in nonparametric regression models when covariates are affected by normal measurement errors. The approach involves applying conditional expectation directly to the kernel-weighted least squares of the discrepancies between the local linear approximation and the observed responses. This innovative algorithm eliminates the need for the simulation step, a characteristic of classical simulation extrapolation methods, thereby significantly reducing computational time. It is worth noting that the method provides an exact form of the extrapolation function. However, obtaining the extrapolation estimate is not as straightforward as merely substituting a negative one for the extrapolation variable in the fitted extrapolation function when the bandwidth is less than the standard deviation of the measurement error. Furthermore, the large sample properties of the proposed estimation procedure are delved into, and simulation studies are conducted. Additionally, a real data example is presented to showcase its practical applications.
