An Introduction to Animal Movement Modeling with Hidden Markov Models using Stan for Bayesian Inference

Abstract

Hidden Markov models (HMMs) are popular time series model in many fields including ecology, economics and genetics. HMMs can be defined over discrete or continuous time, though here we only cover the former. In the field of movement ecology in particular, HMMs have become a popular tool for the analysis of movement data because of their ability to connect observed movement data to an underlying latent process, generally interpreted as the animal's unobserved behavior. Further, we model the tendency to persist in a given behavior over time. Notation presented here will generally follow the format of Zucchini et al. (2016) and cover HMMs applied in an unsupervised case to animal movement data, specifically positional data. We provide Stan code to analyze movement data of the wild haggis as presented first in Michelot et al. (2016).