[Van Pham]

Hi Carlo,

I'm not clear when you said

random effect specification has nothing to do with random selected sample

, so that means I am still able to use random effect model for my panel data even if it is not randomly selected? Or I misunderstood you?

Also I chose the -depvar- based on literature and both Y1 and Y2 could be included. So based on the Hausman test result, do you think I should analyse Y1 with re model while using fe model for Y2.

Thank you so much for your help,

Kind regards,

Van

[Carlo Lazzaro]

Van:
- yes, I meant that you can use random effect model for your panel data even if it is not randomly selected;
- I would go Y1 with -re- specification only.

[Van Pham]

Hi Carlo,

Thank you so much for your information. I will do as you suggested.

Kind regards,

Van