Approximation of Bayes code for Markov sources
Jun’ichi Takeuchi, Tsutomu Kawabata · 2002
We give an approximation formula for the predictive Bayes code for FSMX models (subspaces of Markov models). Moreover, we empirically show that the code using our approximation formula with the Jeffreys prior employed gives shorter code length than the one using the Laplace estimator for the first order Markov models.