st: RE: RE: defining program that is involved with stata and mata interaction

[Nick Cox <n.j.cox@durham.ac.uk>]

The point about different right-quote characters is obscured by one or other of my mailer using plain text and what majordomo does to postings. But different right-quote characters can be distinguished in Zhi Su's original text. 

Nick 
n.j.cox@durham.ac.uk 

Nick Cox

The problem report here is that "it can not get through", "can not work". That is no detail at all on what happens. Also, the program here is incomplete and we can not see your data, so we can't test it for ourselves. All in all, useful comment is made pretty difficult. 

What I can see is that you are using different right quote characters. That won't work. Look at the different characters within 

forvalue k=1/4{
  replace u`k'=0 if u`k'==. &TR==0
}
regress e1 `basic' `T' `u', robust

My guess is that you are using word processing software to edit your code. Bad idea. Use a text editor and use left quote ` and right quote ' to indicate local macros. 

Nick 
n.j.cox@durham.ac.uk 

Zhi Su

I write some commands

  local basic "..."
  gsort -TR
  mvprobit (T1=`basic') (T2=`basic') (T3=`basic') (T4=`basic') if TR==1,robust
  scalar O=e(N)
  mvppred xbm if TR==1,xb

mata

N=3000

O=st_numscalar("O")
.
.

Y=st_data(k,("T1","T2","T3","T4"))

XBM=st_data(k,("xbm1", "xbm2", "xbm3","xbm4"))
.
.
st_store(k,("u1", "u2","u3","u4"),M_c)

end

forvalue k=1/4{
  replace u`k'=0 if u`k'==. &TR==0
}
regress e1 `basic' `T' `u', robust


The program process run well when I run them directly.

Then I want to bootstrapp the whole process. Therefore, I

  capture program drop NHreg
  program define NHreg

  local basic "..."
  gsort -TR
  mvprobit (T1=`basic') (T2=`basic') (T3=`basic') (T4=`basic') if TR==1,robust
  scalar O=e(N)
  mvppred xbm if TR==1,xb

mata

N=3000

O=st_numscalar("O")
.
.

Y=st_data(k,("T1","T2","T3","T4"))

XBM=st_data(k,("xbm1", "xbm2", "xbm3","xbm4"))
.
.
st_store(k,("u1", "u2","u3","u4"),M_c)

end

forvalue k=1/4{
  replace u`k'=0 if u`k'==. &TR==0
}
regress e1 `basic' `T' `u', robust
end

I think the defined command "NHreg" will process all the whole progam.
Then I run command

  NHreg

But the process can not go through at all.

Why the commands can not work through when I put them in "program
define NHreg" and use command "NHreg" to run them?

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/