Simultaneous Switching Regression Model

[Anne Holland]

Hi,

I need help with a command that allows me to do a simultaneous switching regression model with UNKNOWN regime classification (i.e. I don´t know whether the observations come from one regime or another). I want to estimate Demand and Supply simultaneously as per the Porter 1983 Paper "A study on cartel stability".

The issues I have been having are both to do with the fact that it is a simultaneous system of equation, but also with the fact that I do not know the regime. In Porter´s paper, I believe the dummy activates itself endogenously in the model and it also gives rise to different values of the coefficients but I do not know how to do this.

I have tried using switchr but this does not really do the dummy thing mentioned above, and it only gives me the different coefficients in both states but the coefficient on the dummy is a very important thing in Porter´s paper and I wouldn´t be able to get it from switchr - right?

Please let me know if you can help!!

[Phil Bromiley]

You didn't get a quick answer. You'll increase your chances of a helpful answer by following the FAQ on asking questions - provide Stata code in code delimiters, Stata output, and sample data using dataex. Also, try to simplify your code to what is necessary to illustrate your problem.

Don't assume we've reviewed a particular paper. Sometimes folks will do that, but they are more likely to respond to a general question posed in the FAQ format. Someone may yet respond who has studied this article, but normally you get a quick answer if any.

Entering findit switch in the command window brings up a large number of potential estimators. You need to look at each one to see which (if any) do what you want.