Rosenblatt估计与最近邻估计的模拟比较

Simulation study on consistency of Rosenblatt and Nearest neighbors estimate

依据大样本性质,用Monte Carlo方法对特定模型的Rosenblatt估计与最近邻估计进行模拟试验,讨论了在长度有限的样本情况下有关窗宽的选择问题,比较了两种估计的拟合程度及相合速度.

Rosenblatt and Nearest neighbors estimate is the nonparametric estimation of the probability density function. The main work of this article is simulation study on the basis of large sample properties for the estimate by Monte Carlo method with some models. We studied the problem of the choice about window width, consistent rate and some doubtful points about consistent rate with fixed sample size.