论文部分内容阅读
The kernel density estination is the most popular method in density estimation. The main issue is bande width selection，which is a hot topic and is still frustration statisticians，The generalized least aquares cross-validation bandwidth selector is proposed，which significantly improves the classical selector for its variability and undersmoothing,adapts to different kinds of densities，and outperforms the existing bandwidth selectors in statistical literature and software.