TIME SERIES WITH MULTIPLE CHANGE POINTS AND CENSORED OBSERVATIONS

Authors

  • René Castro-Montoya Facultad de Ciencias Físico-Matemáticas, Universidad Autónoma de Sinaloa
  • Gabriel Arcángel Rodríguez-Yam
  • Felipe de Jesús Peraza-Garay
  • José Vidal Jiménez-Ramírez

DOI:

https://doi.org/10.47163/agrociencia.v58i1.2856

Keywords:

Parameter estimation, Bayesian inference, prior distributions, Metropolis algorithm, reversible jump Markov chain Monte Carlo algorithm.

Abstract

This article examines a Bayesian model for a nonstationary time series with an unknown number of change points and censored observations. Each segment is assumed to be an autoregressive process with order one. To estimate the number and locations of change points, we use the reversible jump Markov chain Monte Carlo (RJMCMC) algorithm. The censored problem is solved by imputing the censored values from a multivariate normal distribution based on the observed part. A numerical example shows that the estimates of the number of change points and their localizations have little bias. Additionally, the estimates are robust to the censoring percentage.

Additional Files

Published

22-01-2024

Issue

2024: Early Access (58-1)

Section

Applied Mathematics-Statistics-Computer Science

License

Agrociencia is published every 45 days, in an English format, and it is edited by the Colegio de Postgraduados. Mexico-Texcoco highway Km. 36.5, Montecillo, Texcoco, Estado de México, CP 56264, Telephone (52) 5959284427. www.colpos.mx. Editor-in-Chief: Dr. Fernando Carlos Gómez Merino. Rights Reserved for Exclusive Use: 04-2021-031913431800-203, e-ISSN: 2521-9766, granted by the National Institute for Author Right.