LASSO method in variable selection for right-censored time-to-event data with application to astrocytoma brain tumor and chronic myelogenous leukemia

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

We propose models of potential relationships in time-to-event data and we infer whether these associations occur based on the support of the available data. Our inferences are only as good as the models that we use to define these associations of interest. Cox proportional hazard model [1,2] is the most popular model used in survival analysis due to the computational simplicity of the inference methods and wellestablished asymptotic properties of the partial likelihood. However, the proportional hazards assumption is not always true since the hazard ratio for real data often converges to 1 as time increases.

Original languageEnglish (US)
Title of host publicationDesign and Analysis of Clinical Trials with Time-to-Event Endpoints
PublisherCRC Press
Pages421-440
Number of pages20
ISBN (Electronic)9781420066401
ISBN (Print)9781420066395
Publication statusPublished - Jan 1 2009

    Fingerprint

All Science Journal Classification (ASJC) codes

  • Mathematics(all)
  • Medicine(all)
  • Pharmacology, Toxicology and Pharmaceutics(all)

Cite this

Yu, L., & Pearl, D. K. (2009). LASSO method in variable selection for right-censored time-to-event data with application to astrocytoma brain tumor and chronic myelogenous leukemia. In Design and Analysis of Clinical Trials with Time-to-Event Endpoints (pp. 421-440). CRC Press.