Likelihood Methods in Survival Analysis: With R Examples [Hardcover]

1. Introduction. 2. Semi-parametric Cox Model with Interval Censoring. 3. Extension to Include Truncation. 4.?Extension to Include A Cured Fraction. 5.?Stratified Cox models under interval censoring. 6.?Cox Models with Time-Varying Covariates under Right Censoring. 7.?Copula Cox Models for Dependent Right Censoring. 8.?Additive hazards model. 9.?Parametric Survival Models for Competing Risks Data.

Provides?an overview of the methodology and applications of likelihood methods in survival analysis. It covers all the important topics, including competing risks and joint models. It includes lots of examples to illustrate the methods, and R code for their implementation. A supplementary R package includes all code and data.

Jun Ma, School of Mathematical and Physical Sciences, Macquarie University, North Ryde, Australia

Annabel Webb, School of Mathematical and Physical Sciences, Macquarie University, North Ryde, Australia

Malcolm Hudson, School of Mathematical and Physical Sciences, Macquarie University & NHMRC Clinical Trial Centre, University of Sydney, Sydney, Australia

Many conventional survival analysis methods, such as the Kaplan-Meier method for survival function estimation and the partial likelihood method for Cox model regression coefficients estimation, were developed under the assumption that survival times are subject to right censoring only. However, in practice, survival time observations may include interval-censored data, especially when the exact time of the event of interest cannot be observed. When interval-censored observations are present in a survival dataset, one generally needs to consider likelihood-based methods for inference. If the survival model under consideration is fully parametric, then likelihood-based methods impose neither theoretical nor computational challenges. l“!