On Friday, Anju asked:
>
> I am trying to understand how the predict command
> works after zinb command. I want to know if we can use
> generate command to get the same result as predict
> command.
>
Here is an example that shows what -predict- can do after -zinb-:
clear
webuse fish
drop xb
zinb count persons livebait, inf(child camper) nolog
predict xb1,xb
gen myxb1=[count]_b[persons]*persons /*
*/ +[count]_b[livebait]*livebait+[count]_b[_cons]
sum xb1 myxb1
predict xb2,xb equation(#2)
gen myxb2=[inflate]_b[child]*child /*
*/ +[inflate]_b[camper]*camper+ [inflate]_b[_cons]
sum xb2 myxb2
predict p,p
gen myp=exp(xb2)/(1+exp(xb2))
sum p myp
predict n, n
gen myn=(1-p)*exp(xb1)
sum n myn
My technique for finding out what -predict- is up to behinds the scenes
is to use -set trace on-. Now, a lot of times -set trace on- produces
too much output to search through, but I have found that for the
purposes of finding a -predict- formula, there is surprisingly little
output. Then one can look back through the trace to find the generate
statements that -predict- actually used.
By the way, a nice feature of -set trace on- that I have used, is the
highlight. It highlights a specific word in the output making it easier
to find what you are looking for. Type
whelp trace
to see the help file for that.
-- May
mboggess@stata.com
*
* 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/