BayesReg Version 1.6

Version 1.6 of the BayesReg package has now been released. There are two main changes to this version:

  1. The code now calculates and reports the Widely Applicable Information Criterion (WAIC) score in place of the DIC score.
  2. Options for sampling the coefficients in blocks have been added. This allows the code to be applied to very large predictor matrices (p > 50,000) even on a PC.

Note: The precompiled MEX files for Windows, Linux and MacOSX to increase sampling speed can be obtained here. To use these, all you need to do is download this file and unzip the contents into the “bayesreg” folder.

Bookmark the permalink.

One Response to BayesReg Version 1.6

  1. DH says:

    Hi, Daniel,
    Thanks for publishing the bayesreg package. It is very useful for my research.
    I have a question regarding to Bayesian updating of the model. My problem has 500+ predictor variables, and millions (and growing) observations of very noisy responses. It is not practical to even load all that data in Matlab. So my idea is to break the data into small pieces (say 10,000 observations each), and then run them through bayesreg one by one.
    So instead of using standard prior density provided in baysreg, we will need to feed the posterior density of previous run as prior. Is there a way to do that?
    Thank you very much for your help!

Leave a Reply to DH Cancel reply

Your email address will not be published. Required fields are marked *

*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>