if i understand what you mean, it is the same as with 2 parallel logits.
in these situations, it is usually better, if possible, to stack up both
sets of data in one dataset, include interactions, and then your quantity
of interest (the difference) is a single parameter, which will have a
proper standard error that comes out of the procedure.
Gary
--
*Gary King* - Albert J. Weatherhead III University Professor - Director, IQSS
<http://iq.harvard.edu>- Harvard University
GKing.Harvard.edu <http://gking.harvard.edu/> - King(a)Harvard.edu -
@kinggary<http://twitter.com/kinggary>- 617-500-7570 - Asst 495-9271 -
Fax 812-8581
On Thu, Oct 25, 2012 at 10:12 PM, 笑明 孙 <sun.qqhr(a)yahoo.com.cn> wrote:
Dear Prof. King,
I use relgoit to analyze two different kind of events (e.g. promotion of
male and female workers) with two different dataset (groups).
group 1(all information about male): relogit y1 x1 x2, wc(.###)
group 2(all information about female): relogit y2 x1 x2, wc(.$$$)
Now I want to do a beta difference test to see on which on the two
dependent variables a single independent variable has more impact.
e.g (b1 – b2)
z = --------------------
√(se(b1)2 + se(b2)2)
Can I do that? If yes, what should I need to notice? Suppose the
significance level and sign of x1 and x2 are the same across models.
Any help will be very very appreciated!
Best regards,
Xiaoming Sun
Email: <sun.qqhr@yahoo.com.cn>sun.qqhr@yahoo.com.cn;
sun.413(a)stu.xjtu.edu.cn
Phone No: (+86)15109263565
School of Management
Xi’an Jiaotong University
China
Best regards,
Xiaoming Sun
Email: <sun.qqhr@yahoo.com.cn>sun.qqhr@yahoo.com.cn;
sun.413(a)stu.xjtu.edu.cn
Phone No: (+86)15109263565
School of Management
Xi’an Jiaotong University
China
**
-
---
relogit mailing list served by HUIT
List Address: relogit(a)lists.gking.harvard.edu
Subscribe/Unsubscribe:
http://lists.gking.harvard.edu/mailman/listinfo/relogit
Relogit mailing list
Relogit(a)lists.gking.harvard.edu
To unsubscribe from this list or get other information:
https://lists.gking.harvard.edu/mailman/listinfo/relogit