Assessing the Relative Performance of  Absolute Penalty and Shrinkage Estimation in Weibull Censored Regression Models

Abstract

In this talk, the problem of estimating the vector of regression parameters in the Weibull censored regression model.

Our key objective is to provide natural adaptive estimators that significantly improve upon the classical procedures in the situation where some of the predictors may or may not be associated with the response.  In the context of two competing Weibull censored regression models (full model and candidate sub-model), we consider an adaptive shrinkage estimation strategy that shrinks the full model maximum likelihood estimate in the direction of the sub-model estimate.

Further, we consider  LASSO  strategy and compare the relative performance with the shrinkage estimators. Monte Carlo simulation study reveals that when the true model is close to the candidate sub-model, the shrinkage strategy performs better than the LASSO strategy when, and only when, there are many inactive predictors in the model.

The suggested estimation strategies are applied to a real data set from Veteran's administration  lung cancer study to illustrate the usefulness of the procedures in practice.

Tuesday, October 23, 2012

3:00– 4:00 P.M.

372 Science and Engineering Building

(Refreshments at 2:30-3:00 PM in the kitchen area adjacent to 368 SEB)