I am new to MLE programming and have never done it on Stata before. I am estimating a stochastic production function following Just-Pope (1978).
The formula is the following:
I understand in theory that Stata will maximize the log likelihood function through an iterative method, using the FGLS estimates of the parameters to start. I am, however unable to translate this into code. Previously, I estimated this production function using FGLS, and coded it using loops, but now I'm not sure how I can do this and if I can include fixed effects for instance.
I tried something of the form:
program define maxlike version 14.0 args lnf xb xdelta quietly replace `lnf' = -.5*[_N * ln(2*_pi) + ($ML_y1 - (`xbeta')^2/exp(`xdelta')) + `xdelta'] ml model lf normal (mean: yield_pl*=) (var: yield_pl*=) end
But of course it's not working. Any pointers or references on how I can do this would be greatly appreciated. I should mention that I am working with panel data.
Thank you very much in advance!