matrix tbl = J(2, 1, 0) // collect results to a matrix
matrix colnames tbl = coefficent  // name the rows
matrix rownames tbl = 1 2  // name the column of the matrix
        forv i=1/2{
                qui reg y x if id==`i', noconstant
                matrix tbl[`i', 1] = _b[x]
        }
matrix list tbl // show the results

matrix tbl = J(2, 1, 0) // collect results to a matrix
matrix colnames tbl = coefficent  // name the rows
matrix rownames tbl = 1 2  // name the column of the matrix
        forv i=1/2{
                qui reg y x if id==`i', noconstant
                matrix tbl[`i', 1] = _b[x]
        }
matrix list tbl // show the results
clear
svmat tbl, names(coeff) // use col names as variable names

明明是修正jones模型，感兴趣看我的帖子。。

谢谢大神，但我用这个代码跑出来不是我想要的结果。我想要的是在原来的数据里加入新生成的系数变量，这个要怎么弄呢

set more off
use "D:\file1.dta",clear
matrix tbl = J(2, 2, 0) // collect results to a matrix
matrix colnames tbl = id b  // name the rows
matrix rownames tbl = 1 2  // name the column of the matrix
        forv i=1/2{
                qui reg y x if id==`i', noconstant
                matrix tbl[`i', 1] = `i'
                matrix tbl[`i', 2] = _b[x]
        }
matrix list tbl // show the results
clear
svmat tbl, names(b) // use col names as variable names
rename b1 id
rename b2 b
save "D:\file2.dta",replace

use "D:\file1.dta",clear
merge m:1 id using "D:\file2.dta"
drop _merge