st: smearing factor - same for the sample or group-specific

[Shehzad Ali <sia500@york.ac.uk>]

Hi listers,

Sorry this is probably a non-stata question but any advice would be 
appreciated.

Further to my query about predicting yhat after semi-logarithmic 
regression, I was wondering how should one decide whether to use the same 
smearing factor to predict yhat for the whole sample or different smearing 
factors for different sub-samples of insured and uninsured groups. I read 
somewhere that in RAND study they used different smearing factors for each 
sub-sample of insurance groups.

Any advice on this would be appreciated.

Cheers,

Shehzad
*
*   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/