Dissertation Defense: Sarbesh Pandeya

DrPH Candidate, Biostatistics

Title

Variable Selection in Accelerated Failure Time (AFT) frailty models: An application of Penalized Quasi-Likelihood

Summary

Variable selection is one of the standard ways of selecting models in large scale datasets. It has applications in many fields of research study, especially in large multi-center clinical trials. One of the prominent methods in variable selection is the penalized likelihood, which is both consistent and efficient. However, the penalized selection is significantly challenging under the influence of random (frailty) covariates. It is even more complicated when there is involvement of censoring as it may not have a closed-form solution for the marginal log-likelihood. Therefore, we applied the penalized quasi-likelihood (PQL) approach that approximates the solution for such a likelihood. In addition, we introduce an adaptive penalty function that makes the selection on both fixed and frailty effects in a left-censored dataset for a parametric AFT frailty model. We also compared our penalty function with other established procedures via their performance on accurately choosing the significant coefficients and shrinking the non-significant coefficients to zero.