st: Problems with ML: misclassification-adjusted oprobit

[Ingo Isphording <Ingo.Isphording@rub.de>]

Dear all,

I am trying to implement the misclassification adjustment for binary
choice models proposed by Hausman et al. 1998 via maximum likelihood.
ml check returns an error 133. I am quite a rookie in ml maximization,
so maybe my problem is just a silly typo. So here is my program. Do
you have any suggestions? Many thanks and best regards,

capture program drop mc_prob
program define mc_prob
args lnf beta1 p01 p10
quietly replace `lnf'=ln(`p01'+(1-`p01'-`p10')*normal(`beta1')) if $ML_y1==1
quietly replace `lnf'=ln(1-`p01'-(1-`p01'-`p10')*`beta1') if $ML_y1==0
end

. ml model lf mc_prob (yvar = $expvar, nocons)

. ml check

Test 1:  Calling mc_prob to check if it computes log likelihood and
         does not alter coefficient vector...
         FAILED; mc_prob returned error 133.

Here is a trace of its execution:
------------------------------------------------------------------------------
-> mc_prob __000004 __000003
            - `begin'
            = capture noisily version 9: mc_prob __000004 __000003
              ----------------------------------------------------------------------------------------------
begin mc_prob ---
              - args lnf beta1 p01 p10
              - quietly replace
`lnf'=ln(`a01'+(1-`a01'-`a10')*normal(`beta1')) if $ML_y1==1
              = quietly replace __000004=ln(+(1--)*normal(__000003)) if yvar==1
unknown function +()
              ------------------------------------------------------------------------------------------------
end mc_prob ---
            - `end'
            = set trace off
------------------------------------------------------------------------------
Fix mc_prob.
r(133);
*
*   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/