st: Svyselmlog / Predict Values

[Ana <analunhb@gmail.com>]

Dear Stata users,

I am trying to estimate the wage equations of public and private
sector, while considering the selection that involves the choice of
non-working versus working at public sector versus working at private
sector. I’m following the Bourguignon, Fournier and Gurgand (2004)
model. I’m using the svyselmlog routine (‘selectivity-adjusted
equation based in the multinomial logit for survey data’).

I need to have the predict values of both wage equations (predicted
public sector wage and predicted private wage). And that is the
problem I am facing: I can´t get my predicted values (based on all
regressors,  the dependent variables and  the inverse mills ratio as
well).

Here is the routine:

gen cons = 1
#d ;
svyselmlog lnwhpad exp exp2 fundamental medio superior branco female
NORTE CENTOES SE SUL,select(setgeral exp exp2 fundamental medio
superior branco female NORTE CENTOES SE SUL chefe allincsources filho
idoso)  meth(dmf2)  gen(dmfhpad) force showmlogit;
#d cr
**************
*************
mkmat   exp exp2 fundamental medio superior branco female NORTE
CENTOES SE SUL dmfhpad* cons, matrix(var_rhs)
matrix Bhpad = e(b)
matrix lnwhpad_hat = var_rhs*Bhpad
svmat lnwhpad_hat, names(lnwhpad_predito)
 *************

The Stata’s answer:

**************
mkmat   exp exp2 fundamental medio superior branco female NORTE
CENTOES SE SUL dmfhpad* cons, matrix(var_rhs)
variable dmfhpad* not found
 r(111);
*************

Can anyone help me?
Best regards,
Ana.

*
*   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/