Request: can we have a forecasting unit test that confirms stochastic processes are projected forward as we would expect (e.g. that forecast-period steps in a random walk are generated conditional on the posterior sampled value of the step s.d.)
We do not currently support posterior updates of the step s.d., as far as I can tell. We definitely should, but I think that is out of scope for this PR.
Originally posted by @damonbayer in https://github.com/CDCgov/multisignal-epi-inference/issues/241#issuecomment-2223916492