Change-time detection in moving average model based on reversible jump MCMC algorithm

  • Authors

    • Suparman . Universitas Ahmad Dahlan, Indonesia
    https://doi.org/10.14419/ijet.v7i4.15833

    Received date: July 19, 2018

    Accepted date: March 29, 2019

    Published date: May 27, 2019

  • Bayes Method, Monte Carlo Method, Piecewise Linear, Parameter Estimation, Signal Detection.

  • Abstract

    A piecewise moving average is a model that is more flexible than the moving average model. If the piecewise moving average is used to model data, the model parameters are unknown. The model parameters include the number of models, model coefficients, and white noise variance. This paper discusses the parameter estimation of the piecewise moving average model. Parameter estimation is done using the hierarchical Bayesian approach. Bayes estimators are calculated using the reversible jump Markov chain Monte Carlo algorithm. The performance of the algorithm is tested using synthesis data. The results showed that the reversible jump Markov chain Monte Carlo algorithm can estimate the parameters of the piecewise moving average model well.

  • References

    1. Nakamura, H.; and Toquenaga, Y. (2002). Estimating Colony Loca-tions of Bumble Bees with Moving Average Model. Ecological Re-search, 17(1), 39-48. https://doi.org/10.1046/j.1440-1703.2002.00461.x.
    2. Wang, Y.; and Ding, F. (2017). Iterative Estimation for a Non-Linear IIR Filter with Moving Average Noise by Means of the Data Filter-ing Technique. IMA Journal of Mathematical Control & Information, 34(3), 745-764. https://doi.org/10.1093/imamci/dnv067.
    3. Dimitriou-Fakalou, C. (2011). Yule-Walker Estimation for the Mov-ing Average Model. International Journal of Stochastic Analysis, 1-20. https://doi.org/10.1155/2011/151823.
    4. Diallo, T.M.O.; and Morin, A.J.S. (2015). Power of Latent Growth Curve Models to Detect Piecewise Linear Trajectories. Structural Equation Modeling, 22(3), 449-460. https://doi.org/10.1080/10705511.2014.935678.
    5. Tang, B.; Xiao, Y.; Sivalganathan, S.; and Wu, J. (2017). A Piece-wise Model of Virus-Immune System with Effector Cell-Guided Therapy. Applied Mathematical Modelling, 47, 227-248. https://doi.org/10.1016/j.apm.2017.03.023.
    6. Tang, B.; Xiao, Y.; and Wu, J. (2016). A Piecewise model of Virus-Immune System with Two Thresholds. Mathematical Biosciences, 278, 63-76. https://doi.org/10.1016/j.mbs.2016.06.003.
    7. Xu, Z-H.; Yin, X-A.; Zhang, C.; and Yang, Z.F. (2016). Piecewise Model for Species-discharge Relationships in Rivers. Ecological En-gineering, 96, 208-213. https://doi.org/10.1016/j.ecoleng.2015.12.024.
    8. Hector, M.R-U.; Garnotel, M.; Doron, M.; Jallon, P.; Charpen-Tier, G.; Franc, S.; Huneker, E.; Simon, C.; and Bonnet, S. (2017). An Original Piecewise Model for Computing Energy Expenditure from Accelerometer and Heart Rate Signals. Physiological Measurement, 38(8), 1-1. https://doi.org/10.1088/1361-6579/aa7cdf.
    9. Soleimanmeigouni, I.; Xiao, X.; Ahmadi, A.; Xie, M.; Nissen, A.; and Kumar, U. (2018). Modeling the Evaluation of Ballasted Rail-way Track Geometry by Two-Level Piecewise Model. Structure & Infrastructure Engineering: Maintenance, Management, Life-Cycle Design & Performance, 14(1), 33-45. https://doi.org/10.1080/15732479.2017.1326946.
    10. Hong, A.; and Chen, A. (2012). Piecewise Regression Model Con-struction with Sample Efficient Regression Treen (SERT) and Ap-plications to Semiconductor Yield Analysis. Journal of Process Con-trol, 22(7), 1307-1317. https://doi.org/10.1016/j.jprocont.2012.05.017.
    11. Chamroukhi, F. (2016). Piecewise Regression Mixture for Simulta-neous Functional Data Clustering and Optimal Segmentation. Journal of Classification, 33(3), 374-411. https://doi.org/10.1007/s00357-016-9212-8.
    12. Buscot, M-J.; Wotherpoon, S.S.; Magnussen, C.G.; Juonala, M.; Sa-bin, M.A.; Burger, D.P.; Lehtimaki, T.; Viikari, J.S.A.; Hutri-Kahonen, N.; Raitakari, O.T.; and Thomson, R.J. (2017). Bayesian Hierarchical Piecewise Regression Models: A Tool to detect Trajec-tory Divergence between Groups in Long-Term Observational Stud-ies. BMC Medical Research Methodology, 17, 1-15. https://doi.org/10.1186/s12874-017-0358-9.
    13. Bhansali, R.J. (1983). The inverse partial correlation function of a time series and its applications. J.Multivar. Anal., 13, 310-327. https://doi.org/10.1016/0047-259X(83)90029-5.
    14. Green, P.J. (1995). Reversible Jump MCMC Computation and Bayesian Model Determination. Biometrika, 82(4), 711-732. https://doi.org/10.1093/biomet/82.4.711.
    15. Suparman, (2015). New Segmentation of Piecewise Linear Regres-sion Models Using Reversible Jump MCMC Algorithm. Journal of Computer Technology and Application, 6(1), 14-18. https://doi.org/10.17265/1934-7332/2015.01.003.
    16. Suparman; and Doisy, M. (2014). Hierarchical Bayesian of ARMA Models Using Simulated Algorithm. Journal of Telkomnika, 12(1), 87-96. https://doi.org/10.12928/telkomnika.v12i1.12.

  • Downloads

  • How to Cite

    ., S. (2019). Change-time detection in moving average model based on reversible jump MCMC algorithm. International Journal of Engineering and Technology, 7(4), 6162-6166. https://doi.org/10.14419/ijet.v7i4.15833