KERNEL ESTIMATORS OF GENERAL RADON-NIKODYM DERIVATIVES

Some problems of non parametric inference for point processes use simi lar techniques and produce similar results. The reason for this lies i n the fact that a large family of problems may be regarded in a unifie d way by defining suitable random measures. The object of this paper i s to study kernel estimators using this general approach to estimation . Although the very general setting we will find sufficient conditions for convergence in probability and almost surely, pointwise and unifo rmly, for the kernel estimator. Using this setting vie will also study the associated empirical process.