I'm currently doing my thesis, and i'm having quite some trouble in running the logit regression in stata. If you can help me telling me whether what i'm doing makes sense would be appreciated.

Indeed, I have my database, and what i want to do is to regress a dummy(external, from the 500 most cap firms) over some kind of variables from a sample of firms.

The main problem i'm facing is that, given the external nature of the dependent variable, I don't know how to univocally link it to the dependant variables. As a consequence of that, when i change the order of the dependant variable in the database the results of the regression change dramatically.

For instance what i do is regressing the ext variable over the other variables:

firm.....................A B C D

external dummy..1 0 1 1

Var1....................5 6 7 8

But when i change my database like this:

firm.....................A B C D

external dummy..1 1 0 1

Var1....................5 6 7 8

My results change, is it normal? Do I have to change my way of proceeding or am i doing something wrong?

Thanks a lot

  • 2
    $\begingroup$ When you say your results change but choose not to show us them you do make it rather hard for us to help you. My immediate reaction is that I would have expected to see certain changes but not others so if you can give more details by editing your post someone here might be able to see what is happening. $\endgroup$ – mdewey May 6 at 16:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.