Haskell Stochastic Modeling Tutorial

Question

I would like to use Haskell for stochastic modeling, but I do not know how to do this. I read Hutton's "Haskell Programming," and I find it more convenient to write deterministic functional programs. However, I do not know how to start writing stochastic simulations of this kind that are easy in imperative languages ​​such as R or python. Is there a tutorial or tutorial on this subject that I could read, or can someone give some tips on getting started?

Answer 1

There is a good self-contained paper by Ervig and Colmansberger: Functional pearls - probabilistic functional programming in Haskell on this topic. I used this as a starting point for writing a natural language processor based on Hidden Markov Models in Haskell. There's a package that is based on this article, which also apparently provides a basic interface for building R.

HaskellWiki also has an entry with additional links to hackers. In particular, the ProbabilityMonads package may be useful to you.

Answer 2

http://learnyouahaskell.com/a-fistful-of-monads#the-list-monad

This small section in Learn You the Haskell talks about using the monad and list functor functions to easily deal with non-determinism. It may be a little simplified depending on your needs, but make good use of the tools that are already in the standard library.