Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


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

Re: st: seqlogit with repeated observations


From   Maarten Buis <maartenlbuis@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: seqlogit with repeated observations
Date   Thu, 21 Jun 2012 08:52:10 +0200

On Wed, Jun 20, 2012 at 7:35 PM, wrote:
> I have a query concerning my dataset. I am convinced that I have to use a sequential logit model to address the nature of the dataset, but one stage of the sequence has repeated observations. Hence, I am not sure if using a sequential logit model through the seqlogit function in Stata is the right way to go about my problem. Let me detail the data a little below:
>
> I have a pooled list of listed companies over a period of 10 years. They undergo a decision tree regarding dividend payments. In the first step, the decision involves an agreement with the shareholders to consider dividend payment or not. In the second step, the firm decided to pay dividends or not. This second step can be repeated a number of times in any given year. In the third step, the firm decides to cancel the agreement or continue renewing (annually) the agreement to pay dividends.
>
> I would welcome some insight as to how I can best address this data structure. It seems that the seqlogit function would do the job, but the nature of repeated observations in the second step, where either the number of repeated observations or value of each repeated observation is likely to effect the response variables, make me wary of the nested (sequential) logit model.


-seqlogit- is a user written program (by me) which is available from SSC.

I think that -seqlogit- is too restrictive for modeling the entire
process in one model. I can see here two models: First a sequential
logit, in which you model for the first transition whether or not an
agreement is made, and for the subsequent transitions (one for each
year) whether or not the agreement is canceled. The second model
models for each year whether or not dividends is payed, probably with
the dividend agreement as an explanatory variable.

Hope this helps,
Maarten

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


http://www.maartenbuis.nl
--------------------------

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   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   |   Site index