Page 1 of 1
Heckit starting values
Posted: Wed Apr 29, 2015 9:29 pm
by startz
How does EViews pick starting values for the correlation coefficient when doing a Heckit by mle?
Re: Heckit starting values
Posted: Thu Apr 30, 2015 7:48 am
by EViews Gareth
It uses the estimates calculated as part of the two-step procedure. Essentially ML estimates the full two-step first, and uses those as starting values. Equation 28.43 in EV9 UGII.
Re: Heckit starting values
Posted: Thu Apr 30, 2015 7:51 am
by startz
I see where rho*sigma comes from. Is sigma done from y-x*betaHat from the two-step?
Re: Heckit starting values
Posted: Thu Apr 30, 2015 8:07 am
by EViews Gareth
You can calculate sigma from the SSR of the regression of the response equation, adjusting it for the mills-ratio terms.
sigma = sqrt(SSR/N + theta*theta*mean(lambda))
Re: Heckit starting values
Posted: Thu Apr 30, 2015 8:08 am
by startz
Got it. Thanks.