Dear statalist
I am currently doint analyze of linear almost ideal demand system. I define constraint and then run model by "sureg" command. It didn't work the results show "convariance metrix of error is singular". Could you please give me the idea how I can solve this problem.
Moreover, I would apply Shonkwiler and Yen approach to drawback zero consumption. I already did the probit model and then calculte cdf and pdf from the probit model. What is the stata command to applye these value in the demand system model.
I have five commodity groups which indicate by budget share in each group : w_x (x denotes commodity namely f, v, rg, m, fs) and I have price information for each group which indicate by px. and other household characteristics.
*define constraint for symetry restriction
constraint define 1 [w_f]pv = [w_v]pf
constraint define 2 [w_f]prg = [w_rg]pf
constraint define 3 [w_f]pm = [w_m]pf
constraint define 4 [w_f]pfs = [w_fs]pf
constraint define 5 [w_v]prg = [w_rg]pv
constraint define 6 [w_v]pm = [w_m]pv
constraint define 7 [w_v]pfs = [w_fs]pv
constraint define 8 [w_rg]pm = [w_m]prg
constraint define 9 [w_rg]pfs = [w_fs]prg
constraint define 10 [w_m]pfs = [w_fs]pm
*define constraint for homogeneity restriction
constraint define 11 [w_f]pf+[w_f]pv+[w_f]prg+[w_f]pm+[w_f]pfs = 0
constraint define 12 [w_v]pf+[w_v]pv+[w_v]prg+[w_v]pm+[w_v]pfs = 0
constraint define 13 [w_rg]pf+[w_rg]pv+[w_rg]prg+[w_rg]pm+[w_rg]pfs = 0
constraint define 14 [w_m]pf+[w_m]pv+[w_m]prg+[w_m]pm+[w_m]pfs = 0
constraint define 15 [w_fs]pf+[w_fs]pv+[w_fs]prg+[w_fs]pm+[w_fs]pfs = 0
*Group independent variables in one group
global indep "lnx ln_pf ln_pv ln_prg ln_pm ln_pfs pf pv prg pm pfs edu_y ld size fp awn bkk"
sureg (w_f $indep)(w_v $indep)(w_rg $indep)(w_m $indep)(w_fs $indep), constraints(1 2 3 4 5 6 7 8 9 10 11 12 13 14 15)
Kindly Best Regards,
Rattiya
Rattiya Suddeephong Agricultural Economics (Faculty of Economics) Kasetsart University, Bangkok Thailand Tel: 661-891080466
*
* 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/
