I am trying to run an MCMC chain with a Metropolis-Hastings sampler, but a fixed proposal distribution normally yields poor results. Apparently the AdvancedMH.jl package doesn’t have the feature for tuning the variance of the proposal distribution that is commonly available in MH samplers.
I asked about this features in another topic and it was suggested I opened an issue, but there I didn’t get any response on that. I am not sure if the package is still actively maintained, so instead I am wondering if this could be achieved without actually implementing the missing feature.
My question, therefore, is the following:
- Within the existing implementation of AdvancedMH.jl, is it possible to change the variance of the proposal distribution based on the acceptance rate for the previous
Nsamples?
Thank you.