Closed steve-the-bayesian closed 4 years ago
A crash is never intended, but supporting dynamic regression in this use case is not likely to happen in the near future. Added a check to keep this setup from crashing R, and added a unit test documenting the expected behavior.
I've been recently working with your package "bsts" and found the following problem. If a dynamic regression is included in the state specification and non-consecutive timestamps are provided, R crashes. Is it intended/inevitable or is it a bug?
I attach the example and my sessionInfo.