本人需要处理大批量数据的回归。具体如下：有很多个文件，每个文件格式相同，文件内容可以看做一个矩阵，然后利用每一个文件的某几列数据（经过一定地变换）做一次回归，得到的系数为需要的结果。我已经将很简单却繁琐的命令写了出来，不过还需要手动复制粘贴去取系数，很麻烦。很早前的分组回归的帖子也看了，恐怕解决不了我的问题。望各位高手指教。代码如下：
cap log close
drop _all
insheet using "d:\stata\mydata\000001 (1).txt"
gen x=log((v8+v18)/2)
gen y=log(v5)-x
gen z=x[_n+1]
gen w=z-x
regress w y
drop _all
insheet using "d:\stata\mydata\000001 (2).txt"
gen x=log((v8+v18)/2)
gen y=log(v5)-x
gen z=x[_n+1]
gen w=z-x
regress w y
drop _all
insheet using "d:\stata\mydata\000001 (3).txt"
gen x=log((v8+v18)/2)
gen y=log(v5)-x
gen z=x[_n+1]
gen w=z-x
regress w y
drop _all