st: seqlogit with repeated observations

[rs.net@me.com]

Hello members of Statalist,

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.

Thanks a lot for your help.

Rohit
*
*   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/