Hi! I'm working on understanding the socio-economic factors that dictates access to combination of different credit sources. For MNP (second stage), I have a eight group mutually exclusive classification of the dependent variable that indicates whether the household has borrowed from a combination of sources (E.g.: Only formal, Only MFI, Only Informal, Formal-Informal, MFI-formal etc ).
I wanted to account for non-borrowers in the data. So the first stage is a probit with 1 = Household borrowed and 0 = Not borrowed.
I was trying to use the CMP command in Stata 13 to estimate this in a more efficient manner (as opposed to the 2-stage estimation technique) but I'm getting repeated errors. The code I have used for the same is:
cmp setup
cmp (cate = POOR eduhd gender age asset2 asset3 asset4 asset5 inc2 inc3 inc4 inc5 inc6 inc7 reg2 reg3 grp2 grp3 grp4 grp5 grp6 grp7 psu_sh_cipcag2 psu_sh_cipcag3 psu_sh_mepcag2 psu_sh_mepcag3 psu_sh_snpcag2 psu_sh_snpcag3) (borrow = depend hhsize eduhd eduhd_sq gender age inc2 inc3 inc4 inc5 inc6 inc7 inc_div fam_eve POOR save1 asset2 asset3 asset4 asset5 ), ind(borrow*$cmp_mprobit $cmp_probit)
**ERROR**
Warning: regressor matrix for _mp_cmp_y7 equation appears ill-conditioned. (Condition number = 168.37632.)
This might prevent convergence. If it does, and if you have not done so already, you may need to remove nearly
collinear regressors to achieve convergence. Or you may need to add a nrtolerance(#) or nonrtolerance option to the
> command line.
See cmp tips for more information.
borrow collinear with _cons
I would also like to mention that the same command runs with oprobit but it doesn't work with mprobit.
I tried to do the same with the general example given in the module for women work wage. I see the same thing there. When I replace oprobit with mprobit, the same error comes there. I'm clearly missing out on something very basic. Kindly let me know where I'm going wrong.
Thank you for your time!
I wanted to account for non-borrowers in the data. So the first stage is a probit with 1 = Household borrowed and 0 = Not borrowed.
I was trying to use the CMP command in Stata 13 to estimate this in a more efficient manner (as opposed to the 2-stage estimation technique) but I'm getting repeated errors. The code I have used for the same is:
cmp setup
cmp (cate = POOR eduhd gender age asset2 asset3 asset4 asset5 inc2 inc3 inc4 inc5 inc6 inc7 reg2 reg3 grp2 grp3 grp4 grp5 grp6 grp7 psu_sh_cipcag2 psu_sh_cipcag3 psu_sh_mepcag2 psu_sh_mepcag3 psu_sh_snpcag2 psu_sh_snpcag3) (borrow = depend hhsize eduhd eduhd_sq gender age inc2 inc3 inc4 inc5 inc6 inc7 inc_div fam_eve POOR save1 asset2 asset3 asset4 asset5 ), ind(borrow*$cmp_mprobit $cmp_probit)
**ERROR**
Warning: regressor matrix for _mp_cmp_y7 equation appears ill-conditioned. (Condition number = 168.37632.)
This might prevent convergence. If it does, and if you have not done so already, you may need to remove nearly
collinear regressors to achieve convergence. Or you may need to add a nrtolerance(#) or nonrtolerance option to the
> command line.
See cmp tips for more information.
borrow collinear with _cons
I would also like to mention that the same command runs with oprobit but it doesn't work with mprobit.
I tried to do the same with the general example given in the module for women work wage. I see the same thing there. When I replace oprobit with mprobit, the same error comes there. I'm clearly missing out on something very basic. Kindly let me know where I'm going wrong.
Thank you for your time!
Comment