Skip to content

Research at St Andrews

Markov-switching generalized additive models

Research output: Contribution to journalArticle


Roland Langrock, Thomas Kneib, Richard Glennie, Théo Michelot

School/Research organisations


We consider Markov-switching regression models, i.e. models for time series regression analyses where the functional relationship between covariates and response is subject to regime switching controlled by an unobservable Markov chain. Building on the powerful hidden Markov model machinery and the methods for penalized B-splines routinely used in regression analyses, we develop a framework for nonparametrically estimating the functional form of the effect of the covariates in such a regression model, assuming an additive structure of the predictor. The resulting class of Markov-switching generalized additive models is immensely flexible, and contains as special cases the common parametric Markov-switching regression models and also generalized additive and generalized linear models. The feasibility of the suggested maximum penalized likelihood approach is demonstrated by simulation. We further illustrate the approach using two real data applications, modelling (i) how sales data depend on advertising spending and (ii) how energy price in Spain depends on the Euro/Dollar exchange rate.


Original languageEnglish
Article number1406.3774
Pages (from-to)259-270
Number of pages12
JournalStatistics and Computing
Issue number1
Early online date28 Dec 2015
Publication statusPublished - Jan 2017

    Research areas

  • P-splines, Hidden Markov model, Penalized likelihood, Times series regression

Discover related content
Find related publications, people, projects and more using interactive charts.

View graph of relations

Related by author

  1. Understanding the stochastic partial differential equation approach to smoothing

    Miller, D. L., Glennie, R. & Seaton, A. E., 3 Sep 2019, (Accepted/In press) In : Journal of Agricultural, Biological and Environmental Statistics. In press

    Research output: Contribution to journalArticle

  2. Open population maximum likelihood spatial capture-recapture

    Glennie, R., Borchers, D. L., Murchie, M., Harmsen, B. J. & Foster, R. J., 25 Jul 2019, In : Biometrics. Early View, 11 p.

    Research output: Contribution to journalArticle

  3. Incorporating animal movement into distance sampling

    Glennie, R., Buckland, S. T., Langrock, R., Gerrodette, T., Ballance, L., Chivers, S., Scott, M. & Perrin, W., 8 Sep 2017, (Submitted) In : Journal of the American Statistical Association. 9 p.

    Research output: Contribution to journalArticle

  4. The effect of animal movement on line transect estimates of abundance

    Glennie, R., Buckland, S. T. & Thomas, L., 23 Mar 2015, In : PLoS One. 10, 3, e0121333.

    Research output: Contribution to journalArticle

Related by journal

  1. Statistics and Computing (Journal)

    Michail Papathomas (Reviewer)

    Activity: Publication peer-review and editorial work typesPeer review of manuscripts

Related by journal

  1. Shape constrained additive models

    Pya, N. & Wood, S. N., 2014, (Accepted/In press) In : Statistics and Computing. p. 1-17 17 p.

    Research output: Contribution to journalArticle

  2. Reversible jump methods for generalised linear models and generalised linear mixed models

    Forster, J., Gill, R. & Overstall, A., Jan 2012, In : Statistics and Computing. 22, 1, p. 107-120 14 p.

    Research output: Contribution to journalArticle

  3. Importance tempering

    Gramacy, R., Samworth, R. & King, R., Jan 2010, In : Statistics and Computing. 20, 1, p. 1-7 7 p.

    Research output: Contribution to journalArticle

ID: 119031913