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

st: update seqlogit available on SSC

From   Maarten buis <>
To   stata list <>
Subject   st: update seqlogit available on SSC
Date   Mon, 15 Dec 2008 17:31:54 +0000 (GMT)

Thanks to Kit Baum an update of the -seqlogit- package is now available
on SSC. To install type within Stata: -ssc install seqlogit, replace-,
or -adoupdate-.

-seqlogit- models a process consisting of a sequence of steps or
decisions. For example, the process of attaining education can be seen
as a sequence of steps from one educational level to the next. The 
-seqlogit- package can be used to estimate the influence of explanatory
variables on the probabilities of passing from on level to the next
conditional on being at risk of making that transition. 

This update implements a method of assessing the robustness of the
results against the presence of an unobserved (compound) variable
proposed by Buis (2008). Cameron and Heckman (1998) showed that even if
this unobserved variable is initially not a confounding variable, it
will still result in biased estimates. The -seqlogit- package now has
an option, -sd()-, that will allow you to estimate the effects that
would occur for a given amount of unobserved heterogeneity, that is, a
given standard deviation of the unobserved variable. The idea is to
repeat this for multiple values of this standard deviation and see
whether or not the results change noticeably. If the results change
little for believable values of the standard deviation, then that is an
indication that the results are robust. Moreover, the -seqlogit-
package contains a new program -uhdesc- which describes how the
distribution of the unobserved variable changes over the transitions.
This way one can get an idea how this unobserved variable influences
the results. 

-- Maarten

Buis, Maarten L. "The Consequences of Unobserved Heterogeneity in a
Sequential Logit Model"

Cameron, Stephen V. and James J. Heckman (1998), "Life cycle schooling
and dynamic selection bias: Models and evidence for five cohorts of
american males", The Journal of Political Economy 106(2), 262--333.

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

visiting address:
Buitenveldertselaan 3 (Metropolitan), room N515

+31 20 5986715

*   For searches and help try:

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