How to use HMMs for continuous value prediction

I have some time-series data, which I need to use to predict a continuous value for a given time-stamp. I was initially doing it using a Multivariate Regression Model but I later figured that a time-series based problem could be better solved using Hidden Markov Models.

The dataset consists of a time-stamp label, around 30 features collected from IoT sensors and then there is one target class which is a continuous variable. The problem is how do I determine the structure of the HMM given these features and number of hidden states to represent the high dimensional data and once the above is done, how to predict the continuous target value using the set of transition probabilities and model parameters that get generated.

I have gone through references where the structure of the HMM is specified beforehand which makes it easy to work with, but how do we go about predicting a continuous value with a HMM?

Topic markov-process time-series predictive-modeling machine-learning

Category Data Science

1
answered at

There are continuous extensions to Hidden Markov Model (HMM). They are most commonly called continuous density hidden Markov model (CDHMM).

CDHMM typically assumes a Gaussian distributed observation. The predicted values can often maximum a posteriori probability (MAP) estimates.

About

Geeks Mental is a community that publishes articles and tutorials about Web, Android, Data Science, new techniques and Linux security.