I was wondering if Julia has a Maximum likelihood estimates function like the mle in Matlab.
This function seems to not accept customized PDF, also the package is not 1.0 compatible
I’m pretty sure it’s ready for Julia v1.0.
If you want to do arbitrary ML estimation you would have to use an optimizer like Optim.jl
1 Like
This will allow you to do ML estimation for a user-supplied likelihood function: https://github.com/mcreel/Econometrics.jl/blob/master/src/ML/mleresults.jl
julia> mleresults()
execute edit(mleresults,()) to examine the example code
true betas:
0.60021
0.02946
0.24307
************************************************************
simple ML example
MLE Estimation Results Convergence: true
Average Log-L: -1.29631 Observations: 100
Sandwich form covariance estimator
estimate st. err t-stat p-value
1 0.45890 0.09305 4.93168 0.00000
2 0.08012 0.10026 0.79914 0.42616
3 0.16250 0.10436 1.55706 0.12271
Information Criteria
Crit. Crit/n
CAIC 276.07838 2.76078
BIC 273.07838 2.73078
AIC 265.26287 2.65263
************************************************************
julia>
2 Likes