Good Afternoon,
I am new at Mata but a good user of Stata.
I want to make a GMM with 15 independent variables and 1 dependent.
Problem is that, I have change values of 15 independent variables; so I have 30 independent ones. I have to put in 15 of them. For example X and changeX must not be in the equation at the same time.
First I declared all 30 in stata,
Then
mata
optimize_init_argument(S, 1, X)
CTS =st_data(.,.)
BR=st_data(.,.)
W=cholinv(X'X)
init=J(1,cols(X),0)
S=optimize_init()
optimize_init_evaluator(S, &i_crit())
optimize_init_which(S,"min")
optimize_init_evaluatortype(S,"d0")
optimize_init_params(S,init)
p=optimize(S)
And I receive
initial values not feasible
r(1400);
Please help me.
Thank you in regards,
I am new at Mata but a good user of Stata.
I want to make a GMM with 15 independent variables and 1 dependent.
Problem is that, I have change values of 15 independent variables; so I have 30 independent ones. I have to put in 15 of them. For example X and changeX must not be in the equation at the same time.
First I declared all 30 in stata,
Then
mata
optimize_init_argument(S, 1, X)
CTS =st_data(.,.)
BR=st_data(.,.)
W=cholinv(X'X)
init=J(1,cols(X),0)
S=optimize_init()
optimize_init_evaluator(S, &i_crit())
optimize_init_which(S,"min")
optimize_init_evaluatortype(S,"d0")
optimize_init_params(S,init)
p=optimize(S)
And I receive
initial values not feasible
r(1400);
Please help me.
Thank you in regards,
Comment