Learning Bayesian Statistics cover image

#14 Hidden Markov Models & Statistical Ecology, with Vianey Leos-Barajas

Learning Bayesian Statistics

00:00

How to Select the Number of States?

When you present this to a college, it's not just how many states are there. Are these states meaningful? So kind of have to have this assumption of maybe a normal distribution is meaningful for the patterns that I want to capture. And sort of conditional on our model specification being meaningful, then I think we can get into automatic selection of the numbers of states. But most of the times they might not be meaningful. The big difficulty, even after having thought hard about the scientific knowledge that informs this model is selecting the number of states. That's why I like to use hidden Markov models.

Transcript
Play full episode

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app