[__000000] not found an error occurred when simulate executed my2sls

[Homa Taheri]

Dear Statalists,

I have a problem with the tempvars in the monte carlo simulation of 2sls regression on Stata 13. The "do file" stops working with this error message:


[__000000] not found
an error occurred when simulate executed my2sls
r(111);

I have looked around the web, but almost all the cases were about using the "reshape" command, which is not related to this code. I would appreciate your kind tips. The code is below:


clear all
set seed 10101

* Set the values of the parameters
global numobs = 150 /*numobs = sample size*/
global numsims = 1000 /*numsims = replication number*/

capture program drop my2sls
program my2sls, rclass
version 13
drop _all
set obs $numobs

tempvar x1 x2 x3 eps1 eps2 u1 u2 y1 y2

generate `x1' = runiform()
generate `x2' = runiform()
generate `x3' = runiform()

generate `eps1' = rnormal()
generate `eps2' = rnormal()

generate `u1' = 1.708*`eps1' + 1.404*`eps2'
generate `u2' = 1.732*`eps2'

generate `y1' = -1.412*`x1' - 0.588*`x2' - 2.118*`x3' - 0.588*`u1' - 0.882*`u2'
generate `y2' = -1.412*`x1' - 0.588*`x2' - 2.118*`x3' - 0.882*`u1' - 0.558*`u2'

ivregress 2sls `y2' (`y1' = `x2')

return scalar b1 = _b[`x1']
return scalar b2 = _b[`x2']
return scalar b3 = _b[`x3']

end

set matsize 11000

simulate b1=r(b1) b2=r(b2) b3=r(b3), reps($numsims) saving(results, replace) nolegend nodots: my2sls


use results, clear
summarize



Thanks in advance,
Homa

[daniel klein]

Well you ask Stata to return the coefficients for x1, x2 and x3, yet your model includes only y1 as a predictor and x2 as an instrument. Where do you expect Stata to get the requested coefficients from?

As an aside: try using code delimiters when posting code.

Best
Daniel

[Homa Taheri]

Dear Daniel,

Thank you very much By changing components of the ivregress command, I could solve this problem.



regards,
Homa

[Richard Williams]

In this case the problem was pretty obvious. In other cases it often helps to use the -set trace on- command (perhaps accompanied by -set more off-). There have been plenty of times when I thought a local value had been set when it turned out I had made a mistake somewhere.