Bootstrapping on Implied Timescales

[s-gordon]

Is there currently any inbuilt functionality within MSMB for bootstrapping to assign variance to implied timescales calculated for a particular state representation? I'd like to achieve something similar to what was described here.

If not, has anyone worked up an MSMB-friendly custom script that I can use as a template?

[rmcgibbo]

No, sorry. The next release of MSMBuilder, which is being developed here, includes a BayesianMarkovStateModel with MCMC sampling of the transition matrix posterior, as described in Metzner, Noe, and Schutte (2009). The docs for that are here, and I also made an example notebook, but we haven't started really writing the command line interface yet.