Re: st: Extension of Bivariate Probit

[Muhammad Anees <anees@aneconomist.com>]

What I guess from your discussion,

You need to model:     Y(Fail/Succeed)=f(marketing Options)+Covariates

Is this not a case of simple Logistic Regression with Marketing
Options as Categorical Indicator variables?

So your model should like:      -xt-logit y i.marketing options covariates

where -xt- is optional in case if you had panel data.

Best
Anees

On Tue, Jul 31, 2012 at 3:19 AM, Ayman Farahat <ayman.farahat@yahoo.com> wrote:
> Hello;
> I am working on a modeling whether adopting a certain marketing scheme can lead to a firm's failure.
> The firm has one of n+1 marketing option, where the first n represents the marketing choices and the "n+1" option represents not taking any action. I can model these decisions using a other exogenous covariate.
>
> After adopting any of the "n+1" marketing options, the firm can fail or survive. I can again model the business failure using exogenous covariates.
>
> I want to know whether the two outcomes (choice of marketing AND failure are related). If i only had one marketing decision, i could have used bivariate probit.
> is there an extension of the Bivariate  probit to the case where one of the dependent variables is categorical (not binary and not ordinal)?
> Thanks
> Ayman
>
>
> *
> *   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/



-- 

Best
---------------------------
Muhammad Anees
Assistant Professor/Programme Coordinator
COMSATS Institute of Information Technology
Attock 43600, Pakistan
http://www.aneconomist.com
*
*   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/