Skip navigation

Some hidden Markov models with additional dependence

Some hidden Markov models with additional dependence

Ramesh, Nadarajah (2015) Some hidden Markov models with additional dependence. In: European Meeting of Statisticians (EMS) 2015, July 6, 2015 - July 10, 2015, VU University Amsterdam, Netherlands. (Unpublished)

[img]
Preview
PDF (Author Accepted Abstract)
17977 RAMESH_Some_Hidden_Markov_Models_2015.pdf - Presentation

Download (7kB) | Preview

Abstract

Hidden Markov models can be used to provide a modelling framework for many environmental processes. They can be modified in several ways to accommodate more dependence among observations and this can provide a rich class of flexible models that are useful in many environmental applications. We consider hidden Markov models that incorporate additional dependence among observations in two different ways (Ramesh and Onof, 2014). One approach is to introduce additional dependence between the state level and the observation level of the process and the other is to incorporate dependence at observation level of the process.

Formulation of the additional dependence models and the construction of their likelihood functions, for both approaches, are described. Some special cases of the models and the associated second-order properties of the process are studied. We employ the maximum likelihood methods to estimate the parameters of the models. Proposed models are used to analyse winter season daily rainfall data from England. Results of the analysis show that the models incorporating additional dependence between the state level and observation level of the process capture the structure of the rainfall distribution well whereas the class of models that incorporate dependence at observation level reproduced the autocorrelation structure of the process better than the other models considered.

Reference:
Ramesh, N.I, Onof, C. (2014). A class of hidden Markov models for regional average rainfall. Hydrological Sciences Journal. Vol.59 (9), 1704 -1717.
DOI: 10.1080/02626667.2014.881484

Item Type: Conference or Conference Paper (Paper)
Uncontrolled Keywords: Hidden Markov models, Additional dependence, Rainfall, Maximum likelihood
Subjects: G Geography. Anthropology. Recreation > GE Environmental Sciences
Faculty / Department / Research Group: Faculty of Architecture, Computing & Humanities
Faculty of Architecture, Computing & Humanities > Department of Mathematical Sciences
Last Modified: 10 Nov 2017 12:11
Selected for GREAT 2016: None
Selected for GREAT 2017: None
Selected for GREAT 2018: None
Selected for GREAT 2019: None
URI: http://gala.gre.ac.uk/id/eprint/17977

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year

View more statistics