Page 1 of 1
Too many interaction terms?
Posted: Tue Oct 02, 2012 7:53 am
by Lauw
I'm using a panel data regression to see effects of return on assets on CEO remuneration. Furthermore i've added variables such as board size and board independency. I've also added interaction terms board size*ROA and board independency*ROA, but once i start adding those, ROA falls from significant positive to significant negative. Intuitively, negative ROA-remuneration relation would make little sense. Fixed or random effects change nothing. Is there any way past this besides delete interaction terms?
Re: Too many interaction terms?
Posted: Tue Oct 02, 2012 10:06 am
by startz
Think carefully about what your estimates imply for the partial derivative of remuneration with respect to ROA.
Re: Too many interaction terms?
Posted: Wed Oct 03, 2012 4:54 am
by Lauw
3 ROA terms?
Re: Too many interaction terms?
Posted: Wed Oct 03, 2012 8:26 am
by Lauw
Any way i can get past this then?
Re: Too many interaction terms?
Posted: Wed Oct 03, 2012 8:37 am
by startz
Did you look at the estimated partial derivatives?
Re: Too many interaction terms?
Posted: Thu Oct 04, 2012 5:23 am
by Lauw
How would i do this?
Re: Too many interaction terms?
Posted: Sun Oct 07, 2012 6:12 am
by Lauw
Anybody? I'm pretty stuck.