steve-the-bayesian

The Python interpreter could not find the elpy module.

3

# Summary I'm setting things up on a fresh ubuntu machine. Elpy interrupts frequently when editing python files, so I checked elpy config and got the warning message shown below....

I thought I would bring the following to your attention: 1. It would be nice if we could specify alternative models for the shared factors, other than the random walk...

I fit a model with regressors using the mbsts function (using the long data format), but get the following error message when use the predict method: > fitted$has.regression [1] TRUE...

2. In my understanding there currently is no predict.bsts.mixed. For us it would be nice to have this functionality. We were wondering whether there is no predict for bsts.mixed on...

Some model families leave "sigma" implicit. This has implications for some state models which want to use the "sigma" parameter from the host model. The solution is to separate the...

Add ordinal probit and logit models to BoomSpikeSlab

3

C++ work is mostly done. Just need the glue. They can even use the same posterior sampler.

This will take over much of what BoomSpikeSlab does, but this functionality is needed across packages. void SetRegressionPrior(RegressionModel *model, SEXP r_prior) { if(Rf_inherits(r_prior, "thing1")) { do_thing_1(); } else if(...) {...

Add a continuous time local linear trend.

1

Are there some preferred ways of using bsts for "continuous time" time series i.e. when the intervals between samples are not equidistant? I guess I can time-box the samples and/or...

predict method does not use old.data correctly.

2

Dear Steve, I write you from CIG group of the Technical University of Madrid. We are working in a project where we are trying to apply Bayesian Structural Time Series...

Right now DIRM only supports Gaussian errors. Expand that to include logit, poisson, and student.