How can I deal with xtoprobit 'cannot compute an improvement -- discontinuous region encountered' problem

[Abebe Desta]

how are you doing every one.

I am running "xtoprobit" regression on analysis of Poverty dynamics on 3 round survey data. However, I could not get the regression output. it says discontinuous region encountered. what possible ways are there to solve the problem?

thanks all!

xi: xtoprobit povcategory maried educhead adeqfamsize femmembers deathhhmem illnesshhmem drought flood cropdamage Praisefo
> oditem Incprice_input loss_livestock hh_s12q02 Ftransf_gift inkind_gift inkind_gift animalrent hh_s11aq01 hh_s11aq02 hh_s1
> 1aq03  hh_s14q01 hh_s4q33 i.region, vce(robust)
i.region          _Iregion_1-8        (naturally coded; _Iregion_1 omitted)

Fitting comparison model:

Iteration 0:   log likelihood = -13135.816  
Iteration 1:   log likelihood = -12049.608  
Iteration 2:   log likelihood = -12044.162  
Iteration 3:   log likelihood =  -12044.12  
Iteration 4:   log likelihood =  -12044.12  

Refining starting values:

Grid node 0:   log likelihood = -10095.551

Fitting full model:

Iteration 0:   log pseudolikelihood = -10095.551  
Iteration 1:   log pseudolikelihood = -6245.3878  
adaptive quadrature failed to converge
cannot compute an improvement -- discontinuous region encountered
r(430);

[Carlo Lazzaro]

Abebe:
see reply to the same question at: https://www.statalist.org/forums/for...gression/page2, #16