Skip to main content
. Author manuscript; available in PMC: 2017 Apr 1.
Published in final edited form as: J Neurosci Methods. 2016 Feb 5;263:36–47. doi: 10.1016/j.jneumeth.2016.01.022

Figure 2.

Figure 2

MCMC state trajectories for an HDP-HMM fit to the synthetic dataset shown in Fig. 1. True values are shown by the dotted black lines. The first five iterations of the Markov chain are omitted since they differ greatly from the final states. The chain quickly converges to nearly the correct number of states (A) and achieves close to the true log likelihood (B). (C, D) The chain trajectories of hyperparameters α0 and γ. (E, F) Inferred state transition matrix P and neuronal firing matrix Λ = {λc,i} drawn from the last iteration.