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

RE: st: missing z-values when using dprobit

From   Maarten buis <>
Subject   RE: st: missing z-values when using dprobit
Date   Tue, 27 Nov 2007 07:48:08 +0000 (GMT)

--- "Ohinata, Asako" <> wrote:
> When I try the "noconst" option, however, there are more variables
> with missing standard errors...

This kind of response makes me worried. It looks to me like "lets try
it and see what happens". If you add the -noconst- option the
interpretation of your variables changes drasticly, they are not better
or worse, but if you don't know how to interpret them you will make
grave errors. If you have multiple sets of dummies in interactions you
have multiple reference categories, so restraining one parameter (the
constant) will not solve your problem. There is no quick solution here:
these kinds of models get hopelessly confusing if you do not spend time
carefuly designing your model. This is not difficult, but it does
require time to sit down and write down the comparrisons you want to
make and how this translates into dummies and reference categories, or
other contrasts. See for instance:

-- Maarten 

Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434

+31 20 5986715

Yahoo! Answers - Got a question? Someone out there knows the answer. Try it
*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index