how to use probit model?

[Panos Drosos]

Hello everyone,
i will try to be as concise as possible. My sample consists of two groups of banks. One group with distressed banks and the other with non-distress. I want to chech how some variables (independent) that i have increase or decrease the probability for these banks to go bankrupt(to get distressed). My professor told me that i have to use a probit model but do not split the sample in 2 groups. Questions:1) how can i use the probit regression model in one uniform sample? 2) if i use one uniform sample how i can observe the difrences the differenet effects that my variables have in the 2 groups? Thank you in advance.

[Marcos Almeida]

Welcome to the Stata Forum / Statalist.

I'm not sure whether I understoond your doubt. That said, we don't need to have only 2 groups, for in a probit model we actually have as Yvar "zero" versus "non zero" levels.

[Phil Bromiley]

I think your professor assumed that you would have distressed/not distressed as the dependent variable and the other variables as explanatory variables. You need the distressed and non-distressed in the same sample because you want a model that predicts or explains whether a firm is distressed or not. If you don't have them all together, you would have no variation in your dependent variable and so could not estimate a model.