optimize_couldn't find feasible value

[sungwook kim]

Dear listers,

I'm trying to solve this problem with mata language.Please see the attached.

My code for solving this is as follows.


mata:

mata clear
function linregeval(transmorphic M, real rowvector b, real colvector lnf)
{
real scalar rho
p1 = moptimize_util_xb(M,b,1)
p2 = moptimize_util_xb(M,b,2)
y1 = moptimize_util_depvar(M,1)
y2 = moptimize_util_depvar(M,2)
n =rows(X)
W = st_data(., "splag")
I = I(100)
K = (y11:-(y2:-rho:*p2))
T= -(n:/2):*ln(3.141592):- (n:/2):*ln(1:/n):*K'*K + ln(I:- rho:*W)

lnf = T
}

M = moptimize_init()
moptimize_init_search_random(M, on)
moptimize_init_evaluator(M, &linregeval())
moptimize_init_depvar(M, 1, "test")
moptimize_init_depvar(M, 2, "y1")
moptimize_init_eq_indepvars(M, 1, "wealth")
moptimize_init_eq_indepvars(M, 2, "wealth")
moptimize(M)
moptimize_result_display(M)
end

when I run this code, I get two errors first- conformability- I assume that this is ln(I:- rho:*W), so I removed this part.

but after that I got an error that 'couldn't find feasible values'. How to sort this out?

Any thought would be very welcome!

Kind regards,

Kim