Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Regression Across Two Groups

From   Nick Cox <>
Subject   Re: st: Regression Across Two Groups
Date   Tue, 13 Dec 2011 18:43:04 +0000

I am surprised at the road taken in this thread. To me this is better,
or least more easily, handled as

earnings as response

experience as predictor

foreign qualifications as predictor (indicator variable. some say dummy)

and as such a fairly standard regression model, except that a
logarithmic link or equivalent would seem natural.

I am not an economist, although British readers might like to be
impressed by my A-level in Economics (Grade A) in 1969, so I may
easily be misreading this.


On Tue, Dec 13, 2011 at 5:11 PM, Muhammad Anees <> wrote:

> I know this question is not related directly to Stata, but still
> knowing the fact that an expert advice is possible from my peers in
> statalist. My question is about comparing regression results across
> two groups of individuals where one groups comprises of individuals
> with specific characteristics while the other group does have that
> characteristics. I would be interested in knowing if the relationship
> between any two variables is different across the two groups. For
> example, if I want to test, that the relationship between experience
> and earnings across employees with forign qualifications and without
> forign qualifications is different. I would be thankful to know if
> there is any empirical examples describing such comparisons and how it
> can be accomplished
*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index