| Peer-Reviewed

Generalized Estimation of Missing Observations in Nonlinear Time Series Model Using State Space Representation

Published: 2 April 2013
Views:       Downloads:
Abstract

The aim of the study was to formulate a Time Series Model to be used in obtaining optimal estimates of miss-ing observations. State space models and Kalman filter were used to handle irregularly spaced data. A non-Bayesian ap-proach where the missing values were treated as fixed parameters. Simulated AR (1) data and corresponding estimated missing values were generated using a computer programme. Values were withheld and then estimated as though they were missing. The results revealed that simple exposition of state space representation for commonly used Time Series Models can be formulated.

Published in American Journal of Theoretical and Applied Statistics (Volume 2, Issue 2)
DOI 10.11648/j.ajtas.20130202.13
Page(s) 21-28
Creative Commons

This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.

Copyright

Copyright © The Author(s), 2013. Published by Science Publishing Group

Keywords

Model, Linear, Non-Linear, Simulated, Non-Bayesian

References
[1] Abraham B. and Thaveneswaran, A. (1991). A nonlinear Time Series model and estimation of missing observation. Ann. Inst. Stat.Math Vol 43, 422 – 528.
[2] Abraham B.(1981) Missing observation in time series, Common Statistics A- Theory Methods, 10, 1645-1653.
[3] Ansley, C.F and Kohn, R. (1883). Exact likelihood of vector autoregressive-moving average process with missing aggre-gated data. Biometrika 70, 275-278.
[4] Jones, R.H (1985). Time series analysis with unequally space data. In:E.J. Hannan, P. R.Krishnaiah and M. M. Rao, Eds., Handbook of statistics, vol.5. North-Holland, Amsterdam, 157-177.
[5] Penza, J. and Shea, B. (1997). The exact likelihood of an autoregressive-moving average model with incomplete data. Biometrika 84, 919-928.
[6] Nassiuma. D. (1994). SymmetricStable Sequences with missing observations. J. Time Ser. Anal. 15, 313-323.
[7] Palma, W. and Chan, N. H. (1997). Estimation and forecast-ing of long-memory processes. J. Forecasting 16, 395-410.
[8] Chan,N. H and Palma, W. (1998). State space modeling of long memory process. Ann statist. 26, 719-740.
[9] Palma, W. and Del Pino, G. (1999). Statistical analysis of incomplete long-range dependent data. Biometrika, 86, 4, 965-972.
[10] Engle, R.F. (1982). Autoregressive conditional heterosce-dasticity with estimates of the variance of United Kingdom inflation, Econometrica, 50, 987-1007.
[11] Harrison, P.J. and stevens, C.F. (1976). Bayesian forecasting (with discussion), J. Roy. Statist. Soc.Ser. B, 38, 205-248.
[12] Tjostheim, D (1986). Estimation in nonlinear time series models, stochastic process. Appl. 21, 251-273.
[13] Thavaneswaran, A and Abraham, B. (1988). Estimation for nonlinear timeseries models using estimating equations. J. Time Ser. Anal. 9, 99-108.
[14] Nichols, D.F and B.G. Quinn (1982). Random coefficient Autoregressivemodels: An introduction lecture notes in sta-tistics no. 11. Springer-Verlag, New York.
[15] Ozaki, T. (1985). Nonlinear time series models and dynamic systems, handbook of statistics, vol. 5 (eds, E.J. Hannan, P.R. Krishnaiah and M.M. Rao), 25-83, North Holland, Amster-dam.
[16] Brockwel, P. J. and Davis, R. A. (1987). Time series: Theory and Methods, Springer, New York.
[17] Shiryayev, A.N. (1984). Probability, Graduate test in Math., 95, Springer, New York.
[18] Charbonnier, R., Barlaud, M., Alengrin, G. and Menez, J. (1987). Results on AR- modeling of nonstationary signals, Signal Process., 12, 143-1587.
[19] Miller, R. B. and Ferreiro, O. (1984). A strategy to complete a time series with missing observations, Lecture Notes in statistics, 25, 251-275, Springer, New York.
Cite This Article
  • APA Style

    Biwott K. Daniel, Odongo O. Leo. (2013). Generalized Estimation of Missing Observations in Nonlinear Time Series Model Using State Space Representation. American Journal of Theoretical and Applied Statistics, 2(2), 21-28. https://doi.org/10.11648/j.ajtas.20130202.13

    Copy | Download

    ACS Style

    Biwott K. Daniel; Odongo O. Leo. Generalized Estimation of Missing Observations in Nonlinear Time Series Model Using State Space Representation. Am. J. Theor. Appl. Stat. 2013, 2(2), 21-28. doi: 10.11648/j.ajtas.20130202.13

    Copy | Download

    AMA Style

    Biwott K. Daniel, Odongo O. Leo. Generalized Estimation of Missing Observations in Nonlinear Time Series Model Using State Space Representation. Am J Theor Appl Stat. 2013;2(2):21-28. doi: 10.11648/j.ajtas.20130202.13

    Copy | Download

  • @article{10.11648/j.ajtas.20130202.13,
      author = {Biwott K. Daniel and Odongo O. Leo},
      title = {Generalized Estimation of Missing Observations in Nonlinear Time Series Model Using State Space Representation},
      journal = {American Journal of Theoretical and Applied Statistics},
      volume = {2},
      number = {2},
      pages = {21-28},
      doi = {10.11648/j.ajtas.20130202.13},
      url = {https://doi.org/10.11648/j.ajtas.20130202.13},
      eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ajtas.20130202.13},
      abstract = {The aim of the study was to formulate a Time Series Model to be used in obtaining optimal estimates of miss-ing observations. State space models and Kalman filter were used to handle irregularly spaced data. A non-Bayesian ap-proach where the missing values were treated as fixed parameters. Simulated AR (1) data and corresponding estimated missing values were generated using a computer programme. Values were withheld and then estimated as though they were missing. The results revealed that simple exposition of state space representation for commonly used Time Series Models can be formulated.},
     year = {2013}
    }
    

    Copy | Download

  • TY  - JOUR
    T1  - Generalized Estimation of Missing Observations in Nonlinear Time Series Model Using State Space Representation
    AU  - Biwott K. Daniel
    AU  - Odongo O. Leo
    Y1  - 2013/04/02
    PY  - 2013
    N1  - https://doi.org/10.11648/j.ajtas.20130202.13
    DO  - 10.11648/j.ajtas.20130202.13
    T2  - American Journal of Theoretical and Applied Statistics
    JF  - American Journal of Theoretical and Applied Statistics
    JO  - American Journal of Theoretical and Applied Statistics
    SP  - 21
    EP  - 28
    PB  - Science Publishing Group
    SN  - 2326-9006
    UR  - https://doi.org/10.11648/j.ajtas.20130202.13
    AB  - The aim of the study was to formulate a Time Series Model to be used in obtaining optimal estimates of miss-ing observations. State space models and Kalman filter were used to handle irregularly spaced data. A non-Bayesian ap-proach where the missing values were treated as fixed parameters. Simulated AR (1) data and corresponding estimated missing values were generated using a computer programme. Values were withheld and then estimated as though they were missing. The results revealed that simple exposition of state space representation for commonly used Time Series Models can be formulated.
    VL  - 2
    IS  - 2
    ER  - 

    Copy | Download

Author Information
  • Maseno University,Department of Statistics and Actuarial Science, Kenya

  • Kenyatta University,Department of Statistics, Kenya

  • Sections