next up previous
Next: Wald Test Up: Likelihood Inference for Spatial Previous: Stochastic Approximation

Hypothesis Tests

Consider a test of the null hypothesis that the data in Figure 1.1 is Poisson versus the alternative hypothesis that it comes from a saturation model. Of course, tests of Poissonness are a traditional topic in point processes. The conventional methodology is a nonparametric test using Ripley's K function (see Ripley, 1988, Chapter 3). Here we focus on parametric inference. We calculate the test statistics for the likelihood ratio test and the associated Wald and Rao tests (Rao, 1973, Section 6e3).





Charles Geyer
Fri Jul 5 15:26:21 CDT 1996