st: execution of the heckman command

[Diane Dancer <diane.dancer@sydney.edu.au>]

HI
I have run these models before and had success but can no longer do this.
I have a continuous variable, grade, and a dichotomous variable, hv110.
In the first probit stage I use the hv110 variable and then use the censored observations in the second stage using OLS with the grade variable as the y variable

I have checked the reference manuals to make sure I have the correct y variables in the different commands. 
My commands are as follows

#delimit ;
capture log close;
log using p:\aresearch\jtresearch\Lesotho2004\2011\lesotho080411.log, replace;
use p:\aresearch\jtresearch\Lesotho2004\2011\lesothoddv080411.dta;
summarize hv110 grade;

heckman grade  hv105 hv1052 male foothill mountain kids5 /*
*/hhsize femalehh wealth2 wealth3 wealth4 wealth5 mprim msechigher mnotworking; /*
select(hv110  hv105 hv1052 male urban foothill mountain kids5 /*
*/femalehh wealth2 wealth3 wealth4 wealth5) vce(cluster clus);

I have removed two rows of variables for ease of explanation.
I keep getting this response from Stata   option select() required.

Today, I have imported the data using a text file into Stata and run the program which gives this result.

do "C:\DOCUME~1\dianed\LOCALS~1\Temp\STD01000000.tmp"

. heckman grade = hv105 hv1052 male foothill mountain kids5 /*
> */hhsize femalehh wealth2 wealth3 wealth4 wealth5 mprim msechigher mnotworking
> ; /*
> select(hv110 = hv105 hv1052 male urban foothill mountain kids5 /*
> */femalehh wealth2 wealth3 wealth4 wealth5) vce(cluster clus);

end of do-file



I would be very grateful for any help.
Regards
Diane Dancer

*
*   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/