Statalist The Stata Listserver

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

st: RE: Accessing the t-stat and p-values in a regression

From   "Maarten Buis" <[email protected]>
To   <[email protected]>
Subject   st: RE: Accessing the t-stat and p-values in a regression
Date   Thu, 30 Nov 2006 17:25:22 +0100

--- Nuno Soares wrote:
> I'm running a OLS using the -regression- and -statsby- commands in order to
> save the results into a dataset. Does anyone knows how to access the t-stats
> and p-values associated with the estimated coefficients?
Stata doesn't store the t-stats or the p-values, but you can get the t-stats if
you know the t-stats are the parameter estimates divided by the standard error, 
both of which are stored by -regress-. And once you know the t-stat you can find
the p-value, using the function ttail, see: -help function-. Also see the example 

*----------- begin example -----------------
sysuse auto, clear
regress price weight mpg if foreign == 0
regress price weight mpg if foreign == 1

statsby tweight = (_b[weight]/_se[weight]) /*
     */ pweight = (2 * ttail(e(df_r), abs(_b[weight]/_se[weight]))) /*
     */ tmpg    = (_b[mpg]/_se[mpg]) /*
     */ pmpg    = (2 * ttail(e(df_r), abs(_b[mpg]/_se[mpg]))) /*
     */, by(for): regress price weight mpg
*----------- end example ------------------

Maarten L. Buis
Department of Social Research Methodology 
Vrije Universiteit Amsterdam 
Boelelaan 1081 
1081 HV Amsterdam 
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434 

+31 20 5986715


*   For searches and help try:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index