[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: running multiple regressions at a time and saving coeffs

From   "Nick Cox" <>
To   <>
Subject   st: RE: running multiple regressions at a time and saving coeffs
Date   Wed, 6 Feb 2008 22:16:33 -0000

As Stata is a statistical package, it is very good at computing whatever
descriptive statistics you need. So I see no case for exporting to

On the main part of your question, check out -statsby-. Then whatever
you want is probably just achievable with -summarize-. 


Adrian de la Garza

I have a variable in my dataset called "date," which has approximately
40 different dates in it. I would like to run an OLS regression for each
of these dates, and store all of the coefficients somehow (eg, in a
matrix), to then compute the average and standard deviations of all
these coefficients.

I know I can run all of these regressions by date if I do the following:

bys date: reg y x1 x2 x3

but then this wouldn't save my betas anywhere. So I created a loop as

egen group = group(date)
forvalues i = 1/40 {
reg y x1 x2 x3 if group == `i'
matrix mat`i' = e(b)

So now I have these 40 vectors mat1, mat2, mat3, ... mat40; which I
would like to stack in one big matrix so that I can export it to Excel
or something and compute the descriptive statistics I need, but I don't
know how to do this.

OR, let me know if you know of a better way of doing what I want.

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index