Selecting the number of states in hidden Markov models: pragmatic solutions illustrated using animal movement

DOI10.1007/S13253-017-0283-8zbMATH Open1388.62349arXiv1701.08673OpenAlexW2622738426WikidataQ61465230 ScholiaQ61465230MaRDI QIDQ1680348FDOQ1680348

Authors: Jennifer Pohle, Roland Langrock, Floris M. van Beest, Niels Martin Schmidt 

Publication date: 15 November 2017

Published in: Journal of Agricultural, Biological and Environmental Statistics (Search for Journal in Brave)

Abstract: We discuss the notorious problem of order selection in hidden Markov models, i.e. of selecting an adequate number of states, highlighting typical pitfalls and practical challenges arising when analyzing real data. Extensive simulations are used to demonstrate the reasons that render order selection particularly challenging in practice despite the conceptual simplicity of the task. In particular, we demonstrate why well-established formal procedures for model selection, such as those based on standard information criteria, tend to favor models with numbers of states that are undesirably large in situations where states shall be meaningful entities. We also offer a pragmatic step-by-step approach together with comprehensive advice for how practitioners can implement order selection. Our proposed strategy is illustrated with a real-data case study on muskox movement.

Full work available at URL: https://arxiv.org/abs/1701.08673

Recommendations

• A sequential pruning strategy for the selection of the number of states in hidden Markov models
• Selecting hidden Markov model state number with cross-validated likelihood
• Multi-scale modeling of animal movement and general behavior data using hidden Markov models with hierarchical structures
• Learning hidden Markov models with unknown number of states
• Model selection for hidden Markov chains
• Testing for the number of states in hidden Markov models
• A general hidden state random walk model for animal movement
• Multivariate state hidden Markov models for mark-recapture data
• State space and hidden Markov models

zbMATH Keywords

selection biasinformation criteriaunsupervised learninganimal movement

Mathematics Subject Classification ID

Markov processes: estimation; hidden Markov models (62M05) Applications of statistics to environmental and related topics (62P12) Animal behavior (92D50)

Cites Work

• Estimating the dimension of a model
• Title not available (Why is that?)
• Strictly Proper Scoring Rules, Prediction, and Estimation
• Nonparametric inference in hidden Markov models using P‐splines
• Model Selection and Multimodel Inference
• Bayesian Inference in Hidden Markov Models Through the Reversible Jump Markov Chain Monte Carlo Method
• Hidden Markov models for time series. An introduction using R
• Flexible latent-state modelling of Old Faithful's eruption inter-arrival times in 2009
• Statistical modelling of individual animal movement: an overview of key methods and a discussion of practical challenges
• Multievent: An Extension of Multistate Capture–Recapture Models to Uncertain States
• An introduction to model selection
• Modeling the diving behavior of whales: a latent-variable approach with feedback and semi-Markovian components
• Selecting hidden Markov model state number with cross-validated likelihood
• Title not available (Why is that?)
• A multivariate mixed hidden Markov model for blue whale behaviour and responses to sound exposure

Cited In (39)

• Bayesian inference for continuous-time hidden Markov models with an unknown number of states
• Mining sequences with exceptional transition behaviour of varying order using quality measures based on information-theoretic scoring functions
• Partially hidden Markov chain multivariate linear autoregressive model: inference and forecasting -- application to machine health prognostics
• A hidden Markov model for continuous longitudinal data with missing responses and dropout
• Segmentation of mortality surfaces by hidden Markov models
• Hidden Markov models with applications in cell adhesion experiments
• Hidden Markov Models for Low-Frequency Earthquake Recurrence
• A sequential pruning strategy for the selection of the number of states in hidden Markov models
• A gentle tutorial on accelerated parameter and confidence interval estimation for hidden Markov models using Template Model Builder
• A latent Markov approach for clustering contracting authorities over time using public procurement red flags
• Model-based clustering for noisy longitudinal circular data, with application to animal movement
• Markov-modulated marked Poisson processes for modeling disease dynamics based on medical claims data
• Penalized estimation of flexible hidden Markov models for time series of counts
• Hidden Markov models for multivariate panel data
• Initialization of Hidden Markov and Semi‐Markov Models: A Critical Evaluation of Several Strategies
• Football tracking data: a copula-based hidden Markov model for classification of tactics in football
• Modeling animal movement with directional persistence and attractive points
• Quantile hidden semi-Markov models for multivariate time series
• A Time-Series Model for Underdispersed or Overdispersed Counts
• Emergence of the wrapped Cauchy distribution in mixed directional data
• Penalized composite likelihood estimation for hidden Markov models with unknown number of states
• Copula Modeling of Serially Correlated Multivariate Data with Hidden Structures
• Guest editor's introduction to the special issue on ``Animal movement modeling
• Hidden Markov Models: Discrete Feature Selection in Activity Recognition
• A Markov-switching generalized additive model for compound Poisson processes, with applications to operational loss models
• Integrated population models: achieving their potential
• New formulation of the logistic-Gaussian process to analyze trajectory tracking data
• Bayesian Spline-Based Hidden Markov Models with Applications to Actimetry Data and Sleep Analysis
• Flexible modelling of diel and other periodic variation in hidden Markov models
• A copula-based multivariate hidden Markov model for modelling momentum in football
• The conditionally autoregressive hidden Markov model (CarHMM): inferring behavioural states from animal tracking data exhibiting conditional autocorrelation
• Kernel-based hidden Markov conditional densities
• Model-based two-way clustering of second-level units in ordinal multilevel latent Markov models
• Spline‐based nonparametric inference in general state‐switching models
• Maximum likelihood estimation of multivariate regime switching Student-\(t\) copula models
• A regularized hidden Markov model for analyzing the `hot shoe' in football
• Multi-scale modeling of animal movement and general behavior data using hidden Markov models with hierarchical structures
• Data-driven modeling of the temporal evolution of breakers' states in the French electrical transmission grid
• Hidden semi-Markov-switching quantile regression for time series

Uses Software

• moveHMM

This page was built for publication: Selecting the number of states in hidden Markov models: pragmatic solutions illustrated using animal movement

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1680348)