How to evaluate Ne estimates with different parameters

[djeffares]

Hi all,

I am running smc++ estimate with a non-human species, trying different parameters such as --thinning, --missing_cutoff. the number of distinguished lineages and the number of other lineages.

Is there any parameter in the output files that shows the likelihood of the estimates? I'd like to know which parameters are better for my data?

Also has anyone found that smc++ cv produces more reliable results, for example, compared to some simulated data.

thanks in advance
Daniel Jeffares