A number of methods have been developed for checking the fit of Cox regression models for cohort data; cf. chapter 11 in Klein & Moeschberger (2003, Springer-verlag). However, the methodology for model checking is much less developed when Cox's model is used to analyze nested case-control data. A main reason for this is that the available data in a nested case-control study do not allow for an easy generalization of the common goodness-of-fit methods that are developed for the full cohort. In this thesis we introduce one of the useful model checking techniques for cohort studies, the cumulative residuals process of Lin et al. (1993, Biometrika), and show how it can be extended to nested case-control studies. All formulations regarding nested case-control data are derived and referred to the cohort for comparison. Further, we investigate the performance of model checking for nested case-control studies on real data sets as well as on simulated data, to verify that the method is able to work effectively in various situations.