Thursday, April 9, 2015

Stata embraces Bayesian statistics

Stata 14 has just been released. The new and big thing with version 14 is the introduction of Bayesian Statistics. A wide variety of new models can now be estimated with Stata by combining 10 likelihood models, 18 prior distributions, different types of outcomes, and multiple equation models. Stata has also made available a 255-page reference manual for free to illustrate Bayesian statistical analysis.

Of course R already offered numerous options for Bayesian Inference. It will be interesting to hear from colleagues proficient in Bayesian statistics to compare Stata’s newly added functionality with what has already been available from R.

Given the hype with big data and the newly generated demand for data mining and advanced analytics, it would have been timely for Stata to also add data mining and machine learning algorithms. My two cents: data mining algorithms are in greater demand than Bayesian statistics. Stata users will have to wait for a year or more to see such capabilities. In the meanwhile, R offers several options for data mining and machine learning algorithms.

3 comments:

  1. I use R on a daily basis, but I also like Stata a lot.

    I think you're missing who their target audience is: economists, pharma, etc. (At least that's my impression.) They've recently put a lot of effort into adding and improving Structural Equation Models, and now Bayesian Statistics, and it seems to me that this fits that audience -- and also social science (SPSS) users.

    They are one of the packages that has a growing user base.

    Full-on ML folks will use Matlab, Python, or R, not Stata. Plus Stata (like SAS, SPSS, etc) is command-oriented with a macro/programming language as well, and this dual nature leaves seams that R -- as a full-blown programming language -- doesn't have. Oh, they actually do have some basic ML algorithms if you look in the "Multivariate" manual.

    ReplyDelete
  2. From the introductory materials, it seems that the newly added Bayesian procedures are more or less comparable to what the "MCMCpack" package offers.

    ReplyDelete
  3. Thing is, Stata isn't really meant for that stuff. The median Stata user is an economist doing econometrics.

    ReplyDelete