Cross-Validation in Density Estimation
研究了在有界区间上估计密度时，用交叉验证选择窗宽的方法，发现即使估计量一致，大样本下表现也不佳，建议谨慎使用，并推荐Silverman(1978a)的替代方法。
We examine the cross-validatory method of choosing the window size when estimating a density on a bounded interval. Tt is shown that even if the resulting estimator is consistent it will perform suboptimally for large samples, and should be used cautiously. Alternative techniques, such as that proposed by Silverman (1978a), would be a wiser choice.