MatlabStan

A Matlab interface to Stan, a package for Bayesian inference.

For more information on Stan and its modeling language, see the Stan User's Guide and Reference Manual at http://mc-stan.org/.

Installation

Details can be found in the Getting started page of the wiki.

Example

The following is the classic 'eight schools' example from Section 5.5 of Gelman et al (2003). The output can be compared to that obtained using the Rstan and Pystan interfaces.

schools_code = {
   'data {'
   '    int<lower=0> J;         // number of schools '
   '    real y[J];              // estimated treatment effects'
   '    real<lower=0> sigma[J]; // s.e. of effect estimates '
   '}'
   'parameters {'
   '    real mu; '
   '    real<lower=0> tau;'
   '    real eta[J];'
   '}'
   'transformed parameters {'
   '    real theta[J];'
   '    for (j in 1:J)'
   '    theta[j] = mu + tau * eta[j];'
   '}'
   'model {'
   '    eta ~ normal(0, 1);'
   '    y ~ normal(theta, sigma);'
   '}'
};
  
schools_dat = struct('J',8,...
                     'y',[28 8 -3 7 -1 1 18 12],...
                     'sigma',[15 10 16 11 9 11 10 18]);

fit = stan('model_code',schools_code,'data',schools_dat);

print(fit);

eta = fit.extract('permuted',true).eta;
mean(eta)

A collection of Matlab-specific examples is available in the wiki.