search

steve-the-bayesian / BOOM

A C++ library for Bayesian modeling, mainly through Markov chain Monte Carlo, but with a few other methods supported. BOOM = "Bayesian Object Oriented Modeling". It is also the sound your computer makes when it crashes.
GNU Lesser General Public License v2.1
36 stars 15 forks source link

Standardized prediction errors. #18

Closed steve-the-bayesian closed 6 years ago

steve-the-bayesian commented 6 years ago

Dear Mr. Scott:

I'm using your excellent package bsts and I would like to know if it is possible to compute a standardized version of the one-step-ahead prediction errors - those returned by bsts.prediction.errors() -. My idea is to use them in diagnostic to check homogeneity of variance, outliers, and so on.

Thank you in advance.

Hi Francisco, It could be done. It is conceptually easy, but it probably would take a day or two of work. On the interface side, simply adding a 'standardize' argument to bsts.prediction.errors would do the trick. Then it is just a matter of tracing the changes through the bsts code.

I'll put this on the TODO list, but I can't make any promises about when it will actually get done.

Steve

steve-the-bayesian commented 6 years ago

Done as of 828e584e5ef061582c66d5a4f124b9d9f3fc6d7b.