Hello, [Note: see below for updates, some of the problems here may be resolved] I want to use the Simulator to setup a Bayesian ABC for a large Markov Chain Model. By large lets say it is a state space of couple thousand raised to power of 10 (and not sparsely connected). So the DiscreteMarkovChai…

Using pytensor random number generator with scan for Simulator class

ricardoV94 March 25, 2025, 4:08pm 12

I’m probably be missing something but I would think if you can derive p at each time step you can also marginalize it. Not with pm.marginalize because that assumes a constant transition probability, but the algorithm doesn’t assume that?

1 Like

Topic		Replies	Views
Generating samples using mcmc method from PyMc	20	867	February 27, 2024
Theano.scan and pm.Model() are strangely incompatible Questions	9	1843	May 21, 2019
Trouble with Initilization and Posterior Sampling Questions theano	13	2225	June 15, 2018
Markov switching dynamic autoregression Questions	15	2488	March 8, 2020
Complex model samples very slowly - wondering about possible improvements Questions	5	752	July 31, 2020

Using pytensor random number generator with scan for Simulator class

Related topics