Good evening everyone.
I'm working on Hidden Markov Models and I mainly studied them on the Rabiner tutorial from 1989 and the book "Hidden Markov Models for Time Series: An Introduction Using R, Second Edition" by Zucchini, Macdonald.
I've been able to write on R all the algorithms I need (Forward-Backward, Baum-Welch, Viterbi) by implementing several choices (both discrete and continuous distributions) for the outcome distribution.
I know that it's possible to factor covariates both on the transition probabilities matrix and the prior probabilities. Do you know how I can do it? How should I modify the Baum-Welch algorithm to implement the use of covariates?
Thanks a lot!


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.