Flexible estimation of the state dwell-time distribution in hidden semi-markov models

• Jennifer Pohle ^[1] ; Timo Adam ; Roland Langrock ^[1] ; Larissa Beumer ^[2]
  1. [1] Bielefeld University

     Bielefeld University

     Kreisfreie Stadt Bielefeld, Alemania

     
  2. [2] Aarhus University

     Aarhus University

     Dinamarca

     
• Localización: Proceedings of the 35th International Workshop on Statistical Modelling : July 20-24, 2020 Bilbao, Basque Country, Spain / Itziar Irigoien Garbizu (ed. lit.), Dae-Jin Lee (ed. lit.), Joaquín Martínez Minaya (ed. lit.), María Xosé Rodríguez Álvarez (ed. lit.), 2020, ISBN 978-84-1319-267-3, págs. 189-193
• Idioma: inglés
• Enlaces
  • Texto Completo Libro
• Resumen

  • Hidden semi-Markov models (HSMMs) generalise hidden Markov models by explicitly modelling the time spent in a state, the so-called dwell-time distribution, using some distribution on the positive integers, e.g. the (shifted) Poisson or the negative binomial. In this paper, we propose a penalised maximum likelihood approach for tting HSMMs without the need to specify a distributional assumption for the state dwell times. The feasibility and potential usefulness of the approach is illustrated using muskox animal movement data.