Florian Hartig

Check open branches

7

Hi Tankred / max ... we still have 3 open branches flying around that branched off from development. I think I'll nevertheless merge dev now in master, but can you...

interations = untilConvergence

1

I got a few comments that it would be nice to set iterations until convergence in runMCMC. I guess though that this would require an outer wrapper in runMCMC to...

in particular the samplers

Implement Sensitivity Analysis / Uncertainty tracking on MCMC sampling

1

Make a function to show the sensitivity of a model output on the parameters in the MCMC sample, as well as their contribution to the output uncertainty.

DECRIPTION reference

2

From CRAN: For your next version: Can you provide some references in the 'Description' field of your DESCRIPTION file in the form authors (year) or (no space after 'doi:' and...

How to forward thin argument in the samplers?

6

The thin argument in the samplers is not properly passed on, e.g. ``` library(BayesianTools) ll

Improve SMC

1

At them moment, if iterations = 10, then the likelihood will be divided in 10 equal chunks that are used for the iterative filtering. There was the idea to make...

bayesianSetup / parallel handling

2

The treatment of the parallel option is not properly encapsulated, e.g. ``` tmp

Parallelization of nrChains command

4

Currently, only internal calculations of the MCMC samplers use parallelization, the nrChains call is not automatically parallelized. The current workaround for a slow model, running 3 chains is - start...

Extend possibilities to calculate marginal likelihood

5

Following discussion here http://stats.stackexchange.com/questions/209810/computation-of-the-marginal-likelihood-from-mcmc-samples/210196 see also http://stats.stackexchange.com/questions/44972/what-is-the-difference-between-laplace-approximation-and-modified-harmonic-me