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

Re: st: issues about the consistency of parameter estimation

From   Maarten buis <>
Subject   Re: st: issues about the consistency of parameter estimation
Date   Wed, 28 Oct 2009 08:32:41 +0000 (GMT)

--- On Tue, 27/10/09, amatoallah ouchen wrote:
> My question is about an econometric issue, I'm using
> a  tobit model with friction to estimate 4 parameters, 
> I've already an idea about the value of one of those
> parameters, and I'm using it as a proxy to see whether
> my model gives me good results, but on many cases that
> specific parameter isn't consistent when estimated with
> the tobit model, so I would like to know if this may
> challenge the estimation of the other parameters.

This decision is necessarily a judgement call, so there is
only one peron who can make that decision and that is you.
What helps in making these decisions is whether the 
statistical model imposes that the parameters should be exactly
the same or that the parameters should be similar. In the
former case your results indicate that there is a bug in
one the programs (assuming that inconcistent means a 
difference of more than approx. 1e-4), and that would make
me look again at the programs and at my idea that the model
imposes the constraint that the two should be the exactly same.
Otherwise, I would look if I could explain why the two are
different. Look at what is different between the models 
and look at the data. Try to figure out what kind of 
information from the data is used in each model, and if
there are any patterns in the data that would cause these
differences. Than decide whether these patterns are true 
or artefacts. 

Hope this helps,

Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen


*   For searches and help try:

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