Affordable Access

Model checking techniques for regression models in cancer screening

Authors
Publisher
Collection of Biostatistics Research Archive
Publication Date
Keywords
  • Additive Risk
  • Empirical Process
  • Interval Censoring
  • Regression Diagnostic
  • Right Censoring
  • Clinical Epidemiology
  • Statistical Models
  • Survival Analysis

Abstract

There has been much work on developing statistical procedures for associating tumor size with the probability of detecting a metastasis. Recently, Ghosh (2004) developed a unified statistical framework in which equivalences with censored data structures and models for tumor size and metastasis were examined. Based on this framework, we consider model checking techniques for semiparametric regression models in this paper. The procedures are for checking the additive hazards model. Goodness of fit methods are described for assessing functional form of covariates as well as the additive hazards assumption. The finite-sample properties of the methods are assessed using simulation studies.

There are no comments yet on this publication. Be the first to share your thoughts.