Some of the Most Common Copulas for Simulating Complex Survival Data

Abstract

Simulation studies play an important role in the evaluation of the performance of avariety of statistical methods. Such assessment isperformed under computer intensive proceduresand cannot be achieved with studies of real dataalone. These studies are increasingly employed inevaluating the properties of the proposed methods being the generation of data the most fundamental and important component. However,only a few of published studies provide sufficientdetails to allow readers to understand fully allthe processes to generate the data. In this paperwe present a collection of practical algorithms forsimulating multivariate data from a wide class ofmultivariate copulas. This paper also details important considerations necessary when generating the survival data in a variety of scenarios. Asoftware application for R was developed in whichwe implement all the methods.