hmm-for-behavior | episodes

listen on castbox.fmlisten on google podcastslisten on player.fmlisten on pocketcastslisten on podcast addictlisten on tuninlisten on Amazon Musiclisten on Stitcher


HMM for Behavior

We are joined by Théo Michelot, an assistant professor of Statistics at Dalhousie University, Canada. He works on tracking/time-series data to understand how animals behave.

Théo discussed some of the methods for analyzing data in his field. He discussed how they use random walks to study the movement of animals. He shared the drawbacks of this approach.

Théo explained why Hidden Markov Models (HMMs) are a better approach to modeling animal movement. He discussed collecting acceleration data from animals to understand behavioral transitions. He shared the data variables they typically collect. He shared how they use these variables to define a Hidden Markov Model.

Théo discussed the analysis to check whether the model is accurate. He also shared some of the downsides of HMMs. Théo discussed his R packages for tracking and analyzing moving data.

He discussed the interpretability of HMM and situations where it can be challenging to interpret.

Paper in focus

Understanding the ontogeny of foraging behavior: insights from combining marine predator bio-logging with satellite-derived oceanography in hidden Markov models

Follow our guest


Google Scholar

R Packages




Théo Michelot

I am an assistant professor in statistics at Dalhousie University, Canada, where I started in August 2022. I am part of the Statistical Ecology at Dal (SEaDAL) research group.

Thanks to our sponsors for their support

Georgia Tech Scheller College of Business

Georgia Tech Scheller College of Business

"Are you ready to become a data-savvy leader? Visit to learn more about the Business Analytics Graduate Certificate and apply before the October 1 deadline."