You may think too much about that. According to empirical distribution function, the sample quantiles should be straightforward. note that the quantiles are not 1 to 1 at all. For example, if you want to get the percentile (in 10th) of weight =102.5 from data set sashelp.class... jingju

a=102.5 q=60
NOTE: There were 19 observations read from the data set SASHELP.CLASS.

如果抽样表中有多个数, 可以将抽样表与rank后的原表合并.还可以从rank后的原表直接抽样.

多谢回复！

后来我在R里找到了ecdf函数，可以指定一张表生成经验概率密度函数，然后把这个函数赋给一个类似句柄的东西，然后就可以任意指定观测求分位数了

SAS好像没有类似的函数，是不是要自己写了？谢谢！

高人，学习了。