hi there,
I'm running biprobits and need to calculate out-of-sample
predictions afterwards which can be done by using one of the
predict commands. yet, I also need to calculate a new rho for
these predictions. the initial idea was to use the linear predictions,
calculate the new error terms by simply do y-xbhat and then get
the correlation between them. however, I've tried to do that for
e(sample) predictions in order to replicate the rho given by stata
and the manual rho is by no means identical. is this a
consequence of stata's way to estimate rho in the maximum
likelihood and is there any way to replicate rho manually for out of
sample predictions?
cheers
axel
Axel Heitmueller,
Centre for Economic Reform and Transformation, CERT
School of Management
Heriot-Watt University
Edinburgh
EH14 4AS
UK
phone +44(0)131 451 3969
fax +44 (0)131 451 3296
[email protected]
www.som.hw.ac.uk/somah3/
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
