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: Egen for logit regression

From   muhammed abdul khalid <>
Subject   Re: st: Egen for logit regression
Date   Wed, 7 Apr 2010 17:00:52 +0200

Thank you Martin for the insights. Its good now.

my assumption is that both parents dies, and thus i suspect that the
unexplained premium attached to ethnic1 is due to the inheritance. I
run another regression for those without wealth and find that those
premium disappears.

Thanks again.


2010/4/7 Maarten buis <>:
> --- On Wed, 7/4/10, muhammed abdul khalid wrote:
>> How do i test the effect on ethnicity IF all observations
>> doesn't have inheritances wealth? Do i introduce another
>> dummy say ''inheritance'', of which it equal to wealth>0
>> at age 20? Or use egen command to create mean value of
>> wealth as a proxy of starting point?
> You would need to define what you mean exactly with
> inheritance, in particular does it include transfers from
> living parents to their kids (e.g. to start a business)
> or are you using a more conventional definition which would
> require that (often both) parents died.  If you use the
> latter definition then your first solution would seem wrong:
> fortunately it is very common that the parents are still
> alive when the kids become 20.  If you use the former
> definition, then I would consider these intergenerational
> transfers as an intervening mechanism: they partly explain
> why race influences wealth. In this case I would _not_ want
> to control for these transfers when estimating the effect
> of race. If we would control for these transfers, we would
> take away the part of the effect that we are pretty sure
> really exists (because we know and understand the mechanism)
> and are left with variation we can't explain. Anyhow, this is
> all academic since you appear to have no information on
> intergenerational transfers in your data.
> Hope this helps,
> Maarten
> --------------------------
> Maarten L. Buis
> Institut fuer Soziologie
> Universitaet Tuebingen
> Wilhelmstrasse 36
> 72074 Tuebingen
> Germany
> --------------------------
> *
> *   For searches and help try:
> *
> *
> *


*   For searches and help try:

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