st: mata within the loop does not work

[Oleksandr Shepotylo <shepotylo@gmail.com>]

Dear statalist users,
I have the following code to run 500 Monte Carlo iterations


forvalues i=1(1)500  {
use data, clear
...
mata:
fh = fopen("$p\Geography\weights\weighting_matrix_short", "r")
W=fgetmatrix(fh)
sigma_e=st_numscalar("sigma2")
rho=st_numscalar("rho")
uhat=(I(9)#luinv(I(129)-rho*W))*rnormal(1161, 1, 0, sigma_e)
fclose(fh)
end
...
}




The loop never starts and I get the error r(1). Here is what I got on
the screen:

forvalues i=1(1)100 {
  2.
. use `panel1', clear
  3.
. mata:
  4. fh = fopen("$p\Geography\weights\weighting_matrix_short", "r")
  5. W=fgetmatrix(fh)
  6. sigma_e=st_numscalar("sigma2")
  7. rho=st_numscalar("rho")
  8. uhat=(I(9)#luinv(I(129)-rho*W))*rnormal(1161, 1, 0, sigma_e)
  9. fclose(fh)
 10. end
--Break--
r(1);


It seems that Stata does not want to enter Mata mode. If I remove the
loop and run just one iteration everything works fine. If I take out
the mata part of the code from the loop, the loop runs fine as well. I
also tried using foreach instead and it does not work.

Any help would be greatly appreciated.

Best,
--
Oleksandr Shepotylo
Kyiv School of Economics
Web Page shepotylo.kse.org.ua
Blog shepotylo.blogspot.com

*
*   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/