A Signed-Rank Estimator in Nonlinear Regression Models When Covariates and Errors Are Dependent

This talk will first discuss asymptotic relative efficiency (ARE) of a signed rank estimator in an errors in variables linear regression model with known Gaussian dis tributions of the measurement error, the predicting covariate and its surrogate. The ARE of this estimator relative to the bias corrected least squares estimator at a Gaus sian regression error distribution is shown to increase to infinity as the measurement error variance increases to infinity. Given this motivation, we then derive asymptotic normality of this signed rank estimator in a class of nonlinear semi-parametric regression models where the predicting random covariate vector is possibly dependent on the regression error and where the regression error distribution need not be known.

Co-Author: Palaniappan Vellaisamy, IITB.

Hira L. Koul is Professor Emeritus in the Department of Statistics and Probability at Michigan State University. His areas of research include nonparametric inference, inference on short and long memory processes, time series analysis and survival analysis. He has published around 150 papers, several monographs and guided 35 doctoral theses. His honors included being a Fellow of the American Statistical Association and of the Institute of Mathematical Statistics and Past President of the International Indian Statistical Association. He was a recipient of a Humboldt Research Award for senior scientists in October 1995. He has held visiting positions in numerous national and international universities.

Host: Soumendra Lahiri

This talk is for members the Department of Statistics & Data Science.