RE: st: Statistical Significance of the difference between two estimates from two separate regressions

["Kyrizi, Andri" <ak196@leicester.ac.uk>]

Hello Professor Antonakis,

Thank you so much for your help and for the reference.

The only point I am not sure I understand is: The test of the coefficient b3 is the test of the difference of slopes of the effect of x for males and females. 
--> So the t-test on b3 will be the test of testing the difference?

 I will try this and hopefully I will be able to do it.

All the best,
Andri  
________________________________________
From: owner-statalist@hsphsun2.harvard.edu [owner-statalist@hsphsun2.harvard.edu] on behalf of John Antonakis [John.Antonakis@unil.ch]
Sent: 14 March 2014 10:14
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: Statistical Significance of the difference between two estimates from two separate regressions

Hi:

Two ways to do this:

reg y x if male = 0
est store zero
reg y x if male = 1
est store one
suest zero one

Then see -help suest- on how to test cross-equations coefficients. In
your case it will be:

test x[zero_mean] = x[one_mean]

A more elegant and efficient way to do this is follows (assuming x is
continuous, where we use factor variables, see -help factor variables-):

reg y c.x##i.male

...which will estimate the following:

y = b0 + b1x + b2male + b3x.male + e

The test of the coefficient b3 is the test of the difference of slopes
of the effect of x for males and females.

Of course, for females, the effect of x on y is b1. For males the effect
of x on y is b1 + b3

After estimation, you can see the interaction graphically (suppose the
range of your x-axis is 1 through 20:

margins male, at(x = (1 20))

To better understand interactions see:

Aiken, L. S. & West, S. G. 1991. Multiple Regression: Testing and
Interpreting Interactions. Newbury Park, CA: Sage Publications.

HTH,
J.

__________________________________________

John Antonakis
Professor of Organizational Behavior
Director, Ph.D. Program in Management

Faculty of Business and Economics
University of Lausanne
Internef #618
CH-1015 Lausanne-Dorigny
Switzerland
Tel ++41 (0)21 692-3438
Fax ++41 (0)21 692-3305
http://www.hec.unil.ch/people/jantonakis

Associate Editor:
The Leadership Quarterly
Organizational Research Methods
__________________________________________

On 14.03.2014 11:03, Kyrizi, Andri wrote:
> Dear Statalisters,
>
> I am running two (pooled ols) wage regressions: one for males and one for females.
>
> I would like to test whether there is a difference between the estimates of the two groups and if the difference is statistically significant.
>
> Most importantly I am interested to see if the coefficient I receive for education is statistically different between the two groups.
>
> Could anyone help me with this? Is there a test to do this?
>
> Thank you,
> Andri
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/