Markov-switching generalized additive models

Roland Langrock, Thomas Kneib, Richard Glennie, Théo Michelot

Research output: Contribution to journalArticlepeer-review

21 Citations (Scopus)

Abstract

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
Volume27
Issue number1
Early online date28 Dec 2015
DOIs
Publication statusPublished - 1 Jan 2017

Keywords

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

Fingerprint

Dive into the research topics of 'Markov-switching generalized additive models'. Together they form a unique fingerprint.

Cite this