Getting errors/standard deviation from OptimizationResults object

[astro-kevin]

Is there a way to get the error/standard deviation of parameters from a OptimizationResults object?

[robertfeldt]

Standard deviation of what? In the parameter positions? If you are using a population-based optimizer it has a population of solutions so if you wanted you could get all those solutions and study the std over the parameters. But very unclear what the value would be since low std could just indicate that the population has converged to almost the same optimum. This might not say much about other optima. Can you describe more what you are trying to achieve?

[astro-kevin]

I guess what I'm looking for is the inverse of the hessian matrix returned by the fit because I'm fitting a negative log likelihood function, and that'll get me the gaussian approximation to the maximum likelihood. I know that such a thing exists in python packages which I used before, but I don't know if there's a way to get this here.

[robertfeldt]

This is in general not available from heuristic, black-box optimization algorithms. The whole point with them is that they can be applied even when the derivatives or the Hessian cannot be calculated or isn't efficiently available. So you might be better off investigating the Optim.jl package, see for example: https://julianlsolvers.github.io/Optim.jl/stable/#examples/generated/maxlikenlm/

[astro-kevin]

Thanks, I guess I didn't understand how BlackBoxOptim worked.