Testing Goodness-of-Fit of Parametric Survival Models for Right Censored Data

Abstract

The main goal of this work it is to present a review of the existing methods to deal with the goodness-of-fit for right-censored data. Goodness-of-fit tests are developed to assess whether a given distribution is suited to a data set. Literature on goodness-of-fit tests for right-censored data is scarce and scattered. This master s degree thesis is divided into three different parts. The first part is devoted to review the bibliography of goodness-of-fit test for parametric models with right-censored data. We classify them according to the type of censoring and the methodology used, and we also propose a unified notation. The second part it focuses on the theoretic explanation of the Generalized Chi Squared test. Finally, the last part of the work presents an implementation in R of the Generalized Chi-Squared test for complete and right-censored data. We also have applied the above methods to some data sets and we have analyzed the results.