Statalist The Stata Listserver


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

st: Oprobit, IV, clustering and weights


From   <J.Fischer@lse.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: Oprobit, IV, clustering and weights
Date   Wed, 3 May 2006 18:20:05 +0100

Dear all

I would like to instrument one variable in an ordered probit regression.
My sample contains individual data with sample weights, combined with an
institutional variable. The to be instrumented variable is at the
aggregate level.

So for my (uninstrumented) analysis I use weights (pw) and cluster by
the aggregate level(Moulton 1990).

A 'first stage regression' gives me nicely predicted values.  (I
selected the instruments using ivreg2). 

My question is
* how can I take into account the individual weights in the second stage
? The vce(boot) option does not allow the usage of weights. 
(or I this not crucial in this case ?)

* are boostrapped standard errors the right choice in my case ? ( my
knowledge is very limited in this area)

* bootstrapping,stata rops all my aggregate variables ( I have a sample
of 24) independently of whether I use the cluster option or not


Thanks

Justina



*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index