I'm using jags to fit a model for a problem similar to the "matrix factorization". This means that the data I provide to the system can be summarized in a very sparse matrix.
Now, because this matrix is large (1000*1000 more or less), the system require a huge number of iterations before the adaptation is complete and a lot of time for each iteration. Exist a way to menage this kind of dataset and to speed up the MCMC phase??