Forecasting with Pairwise Gaussian Markov Models

Marc Escudier; Ikram Abdelkefi; Clement Fernandes; Wojciech Pieczynski

doi:10.1109/MCSI60294.2023.00009

Forecasting with Pairwise Gaussian Markov Models

Marc Escudier, Ikram Abdelkefi, Clement Fernandes, Wojciech Pieczynski

Telecom Sudparis

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Pairwise Markov Models (PMMs) extend the well-known Hidden Markov Models (HMMs). Being significantly more general, PMMs enable several types of processing, like Bayesian filtering or smoothing, similar to those used in HMMs. In this paper, we deal with Bayesian forecasting. The aim is to show analytically in the simple stationary Gaussian case that the extent results obtained with HMM can be improved. We complete contributions with a theoretical error study and two real examples we deal with. Experiments show that PMMs-based forecasting can significantly improve HMMs-based ones.

Original language	English
Title of host publication	Proceedings - 2023 8th International Conference on Mathematics and Computers in Sciences and Industry, MCSI 2023
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1-8
Number of pages	8
ISBN (Electronic)	9798350341652
DOIs	https://doi.org/10.1109/MCSI60294.2023.00009
Publication status	Published - 1 Jan 2023
Event	8th International Conference on Mathematics and Computers in Sciences and Industry, MCSI 2023 - Athens, Greece Duration: 14 Oct 2023 → 16 Oct 2023

Publication series

Name	Proceedings - 2023 8th International Conference on Mathematics and Computers in Sciences and Industry, MCSI 2023

Conference

Conference	8th International Conference on Mathematics and Computers in Sciences and Industry, MCSI 2023
Country/Territory	Greece
City	Athens
Period	14/10/23 → 16/10/23

Keywords

Gaussian models
Pairwise Markov models
forecasting
hidden Markov models

Access to Document

10.1109/MCSI60294.2023.00009

Cite this

Escudier, M., Abdelkefi, I., Fernandes, C., & Pieczynski, W. (2023). Forecasting with Pairwise Gaussian Markov Models. In Proceedings - 2023 8th International Conference on Mathematics and Computers in Sciences and Industry, MCSI 2023 (pp. 1-8). (Proceedings - 2023 8th International Conference on Mathematics and Computers in Sciences and Industry, MCSI 2023). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/MCSI60294.2023.00009

Escudier, Marc ; Abdelkefi, Ikram ; Fernandes, Clement et al. / Forecasting with Pairwise Gaussian Markov Models. Proceedings - 2023 8th International Conference on Mathematics and Computers in Sciences and Industry, MCSI 2023. Institute of Electrical and Electronics Engineers Inc., 2023. pp. 1-8 (Proceedings - 2023 8th International Conference on Mathematics and Computers in Sciences and Industry, MCSI 2023).

@inproceedings{39e031953a7c40f9ba991bb439197a24,

title = "Forecasting with Pairwise Gaussian Markov Models",

abstract = "Pairwise Markov Models (PMMs) extend the well-known Hidden Markov Models (HMMs). Being significantly more general, PMMs enable several types of processing, like Bayesian filtering or smoothing, similar to those used in HMMs. In this paper, we deal with Bayesian forecasting. The aim is to show analytically in the simple stationary Gaussian case that the extent results obtained with HMM can be improved. We complete contributions with a theoretical error study and two real examples we deal with. Experiments show that PMMs-based forecasting can significantly improve HMMs-based ones.",

keywords = "Gaussian models, Pairwise Markov models, forecasting, hidden Markov models",

author = "Marc Escudier and Ikram Abdelkefi and Clement Fernandes and Wojciech Pieczynski",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 8th International Conference on Mathematics and Computers in Sciences and Industry, MCSI 2023 ; Conference date: 14-10-2023 Through 16-10-2023",

year = "2023",

month = jan,

day = "1",

doi = "10.1109/MCSI60294.2023.00009",

language = "English",

series = "Proceedings - 2023 8th International Conference on Mathematics and Computers in Sciences and Industry, MCSI 2023",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1--8",

booktitle = "Proceedings - 2023 8th International Conference on Mathematics and Computers in Sciences and Industry, MCSI 2023",

}

Escudier, M, Abdelkefi, I, Fernandes, C & Pieczynski, W 2023, Forecasting with Pairwise Gaussian Markov Models. in Proceedings - 2023 8th International Conference on Mathematics and Computers in Sciences and Industry, MCSI 2023. Proceedings - 2023 8th International Conference on Mathematics and Computers in Sciences and Industry, MCSI 2023, Institute of Electrical and Electronics Engineers Inc., pp. 1-8, 8th International Conference on Mathematics and Computers in Sciences and Industry, MCSI 2023, Athens, Greece, 14/10/23. https://doi.org/10.1109/MCSI60294.2023.00009

Forecasting with Pairwise Gaussian Markov Models. / Escudier, Marc; Abdelkefi, Ikram; Fernandes, Clement et al.
Proceedings - 2023 8th International Conference on Mathematics and Computers in Sciences and Industry, MCSI 2023. Institute of Electrical and Electronics Engineers Inc., 2023. p. 1-8 (Proceedings - 2023 8th International Conference on Mathematics and Computers in Sciences and Industry, MCSI 2023).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Forecasting with Pairwise Gaussian Markov Models

AU - Escudier, Marc

AU - Abdelkefi, Ikram

AU - Fernandes, Clement

AU - Pieczynski, Wojciech

PY - 2023/1/1

Y1 - 2023/1/1

N2 - Pairwise Markov Models (PMMs) extend the well-known Hidden Markov Models (HMMs). Being significantly more general, PMMs enable several types of processing, like Bayesian filtering or smoothing, similar to those used in HMMs. In this paper, we deal with Bayesian forecasting. The aim is to show analytically in the simple stationary Gaussian case that the extent results obtained with HMM can be improved. We complete contributions with a theoretical error study and two real examples we deal with. Experiments show that PMMs-based forecasting can significantly improve HMMs-based ones.

AB - Pairwise Markov Models (PMMs) extend the well-known Hidden Markov Models (HMMs). Being significantly more general, PMMs enable several types of processing, like Bayesian filtering or smoothing, similar to those used in HMMs. In this paper, we deal with Bayesian forecasting. The aim is to show analytically in the simple stationary Gaussian case that the extent results obtained with HMM can be improved. We complete contributions with a theoretical error study and two real examples we deal with. Experiments show that PMMs-based forecasting can significantly improve HMMs-based ones.

KW - Gaussian models

KW - Pairwise Markov models

KW - forecasting

KW - hidden Markov models

U2 - 10.1109/MCSI60294.2023.00009

DO - 10.1109/MCSI60294.2023.00009

M3 - Conference contribution

AN - SCOPUS:85190618878

T3 - Proceedings - 2023 8th International Conference on Mathematics and Computers in Sciences and Industry, MCSI 2023

SP - 1

EP - 8

BT - Proceedings - 2023 8th International Conference on Mathematics and Computers in Sciences and Industry, MCSI 2023

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 8th International Conference on Mathematics and Computers in Sciences and Industry, MCSI 2023

Y2 - 14 October 2023 through 16 October 2023

ER -

Escudier M, Abdelkefi I, Fernandes C, Pieczynski W. Forecasting with Pairwise Gaussian Markov Models. In Proceedings - 2023 8th International Conference on Mathematics and Computers in Sciences and Industry, MCSI 2023. Institute of Electrical and Electronics Engineers Inc. 2023. p. 1-8. (Proceedings - 2023 8th International Conference on Mathematics and Computers in Sciences and Industry, MCSI 2023). doi: 10.1109/MCSI60294.2023.00009

Forecasting with Pairwise Gaussian Markov Models

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this