Page 1 of 1

Inverse Mills Ratio

Posted: Wed Oct 09, 2013 1:00 am
by barryoli
Hi,
I'm trying to calculate the IMR...I've looked at the response by Gareth:
eq01.makeresid ordinary
eq01.makeresid(g) general
eq01.fit yprob
series imills2 = general*(1-yprob)/ordinary

I can understand creating the ordinary residuals and the generalized residuals...done that.
I can understand creating the series imills2=....

where i'm stuck is what this means "eq01.fit yprob"
or how to calculate it from my model...

my equation is:
fin= c firm_age size growth ...
(where fin= binary for firm seeking finance and the independent variables include firm age, size growth etc...)...saved both ordinary and general residuals...but stuck on the "eq01.fit yprob"...

the next stage is to estimate:

loan= c firm_age size growth [to see what variable is significant for credit rationing] and add the IMR as the final independent variable.

loan is the % of loan applied and since its only a subset of those who have sought finance there's self selection bias.


Any help appreciated...


Barry.

Re: Inverse Mills Ratio

Posted: Wed Oct 09, 2013 11:02 am
by EViews Glenn

Code: Select all

eq01.fit yprob
is just a proc off of EQ01 that computes the fitted values of the probabilities.

Re: Inverse Mills Ratio

Posted: Wed Oct 09, 2013 2:16 pm
by barryoli
Thanks for the quick response...I think I was a bit tired last night when doing this stuff...but all is good.

Cheers
Barry.