bayesglm in Matlab?

Posted on November 6, 2008 10:34 PM by Andrew

I received the following email:

I have just read your paper “A default prior distribution for logistic and other regression models”.
It looks very interesting and useful for general bayesian models. . . . However, as an engineer, I know nothing about R and use MATLAB for everything! I don’t suppose you have the algorithm available in MATLAB code?

Any thoughts? I don’t think it should be too hard to translate from R to Matlab; unfortunately bayesglm() is a hack on glm() which is full of irrelevant code (exception-handling, etc). It would be easier to program from scratch in Matlab, I think.

3 thoughts on “bayesglm in Matlab?”

Jarad on November 7, 2008 1:58 AM at 1:58 am said:

Learn R?

Seriously, it shouldn't be too hard if you already know MATLAB. Especially if all you want to do is read in data from a csv file, run bayesglm, and output the results to another csv file.
Frank DiTraglia on November 7, 2008 12:43 PM at 12:43 pm said:

I've found this bilingual R-MATLAB manual really helpful:

http://www.math.umaine.edu/faculty/hiebeler/comp/…
Kjetil Halvorsen on November 10, 2008 5:18 AM at 5:18 am said:

Learning R is probably the best solution, but I don't by into "learning R is easy once you know matlab". Before R was around I used matlab a lot, and it took me exactly 5 min to learn matlab! I know of nobody who learnt R in 5 min. R is a (sometimes unreasonably complicated) functional language, matlab is simply a simplified and improved pascal. matlab is fast to learn because there is hardly anything to learn at all. You dont need to learn OO programming, it isn't there. Nothing to learn, on the other hand: no help to get!!

Comments are closed.