Finite sample moments of a bootstrap estimator of the james-stein rule
推导了詹姆斯-斯坦规则自助法估计量的有限样本矩，发现其有偏，并分析了偏误来源，提示自助法在类似非线性情形下也可能有偏。
The finite sample moments of the bootstrap estimator of the James-Stein rule are derived and shown to be biased. Analytical results shed some light upon the source of bias and suggest that the bootstrap will be biased in other settings where the moments of the statistic of interest depends on nonlinear functions of the parameters of its distribution.