ABSTRACT
In this chapter we discuss nonparametric estimation methods that can be used to describe the characteristics of the process under study. Because these methods do not make any assumptions about the distribution of the process, they are particularly suited for first exploratory data analyses. TDA contains procedures to calculate life tables and Kaplan-Meier (or product limit) estimates. Both of these methods are helpful for graphical presentations of the survivor function (and their transformations) as well as the transition rate. The life table method is the more traditional procedure and has been used in the case of large data sets because it needs less computing time and space. However, compared to the Kaplan-Meier estimator, the life table method has the disadvantage that the researcher has to define discrete time intervals, as is shown later. Given modern computers, there seems to be no reason anymore to prefer the life table method on the basis of computer time or storage space. We therefore give only a few examples for the life table method and discuss the Kaplan-Meier estimator in more detail.
