之前一位同学问及bootstrap，由此我查阅了几篇文献，初步知晓个皮毛：它是一种非参检验方法，利用重复抽样理论，来减少偏差、控制方差、得到有效置信区间等统计方法。国内bootstrap研究比较少，这里摘录了国外研究成果。

1.what is bootstrapping

The method of bootstrapping was first introduced by Efron as a method to derive the estimate of standard error of an arbitrary estimator.

The basic idea behind a bootstrap is that for some reason we do not know how to calculate the theoretically appropriate significance test for a statistic: Some possible examples are that we want to do a t-test on a mean when the data is nonnormal or perhaps we want to do a t-test on a median or maybe we want do a regression where the assumptions about the error term are violated.

Using the bootstrap approach assumes that the data are a random sample. The bootstrap simulates what would happen if repeated samples of the population could be taken by taking repeated samples of the data available. These repeated samples could each be smaller than the data available and could be done with or without replacement.

Empirical research suggests that the best results are obtained with the repeated samples are the same size as the original sample and when it is done with replacement.    (《A Practical Introduction to the Bootstrap Using the SAS System》)

bootstrap方法的实践需要借助计算机语言，人工计算显得无力。关键是对重复抽样的控制不易操作，比较复杂。

下面是sas实现的例子：

2.performing bootstrap analyses in sas

 Resampling using data and sort steps   （《a tuitorial on bootstrapping in the sas system》）