Dear all,
I am working on creating some code that performs a parameter fit for a binary black hole merger (GW150914). I am using Bilby for this; specifically, I am using the Dynesty sampler. However, I am having issues with the convergence of the sampler. The sampler simply does not converge and does not produce any results. I tried copying and pasting the tutorial (see: Compact binary coalescence parameter estimation — bilby 2.3 documentation) but even this code does not produce the outcome that is listed below the code.
Looking at the logs, I see that the logarithm of the evidence varies greatly, which obviously is an issue. I am using priors specifically designed for event GW150914, so I know that the priors are not the problem.
I tried tweaking various parameters of the sampler but have not yet succeeded in solving the issue. If anyone here is familiar with convergence issues and has any pointers on what parameters I could look into, it would be greatly appreciated.
I’d be happy to share any information from the logs of my failed runs, if necessary. I hope someone is able to point me in the right direction.
All the best,
Wout.