Intervention Analysis in Multivariate Time Series via the Kalman Filter

  • David K. Blough
Part of the Lecture Notes in Statistics book series (LNS, volume 55)


The Kalman filter has been found useful in fitting ARIMA models. This paper uses the multivariate extension of the state-space approach to fit multivariate time series which include covariates and periodic interventions. An example will be presented in which these techniques are applied to the study of the pink cotton boll worm moth, Pectinophora gossypiella (Saunders), (Lepidoptera: Gelechiidae). Moth counts were taken at various locations in a large cotton field over time. A special type of multivariate time series will be used, namely a spatial time series. The amount of irrigation water at each location will be included in the model as a covariate and periodic applications of insecticide will be included as interventions. Previously developed techniques for handling missing data, aggregate data, and nonlinear data transformation are also incorporated. Maximum likelihood estimates of the model parameters are obtained by imbedding the filter in a quasi-Newton optimization routine.


Irrigation Water Kalman Filter Time Series Analysis Time Series Model ARMA Model 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Anderson, B. D. O. & J. B. Moore. 1979. Optimal Filtering. Englewood Cliffs: Prentice-Hall.zbMATHGoogle Scholar
  2. Ansley, C. F. & R. Kohn. 1983. Exact likelihood of a vector autoregressive-moving average process with missing or aggregate data. Biometrika 70: 275–278.MathSciNetzbMATHCrossRefGoogle Scholar
  3. Bard, Y. 1974. Nonlinear Parameter Estimation. New York: Academic Press.zbMATHGoogle Scholar
  4. Box, G. E. P. & G. M. Jenkins. 1970. Time Series Analysis, Forecasting, Control. San Francisco: Holden-Day.Google Scholar
  5. Box, G. E. P. & G. C. Tiao. 1975. Intervention analysis with applications to economic and environmental problems. J. Amer. Statist. Assoc. 70: 70–79.MathSciNetzbMATHCrossRefGoogle Scholar
  6. Gardner, G., A. C. Harvey & G. D. A. Phillips. 1980. An algorithm for exact maximum likelihood estimation of autoregressive-moving average models by means of kalman filtering. Appi. Statist 29: 311–322.zbMATHCrossRefGoogle Scholar
  7. Gill, P. E. & W. Murray. 1972. Quasi-Newton methods for unconstrained optimization. J. Inst. Math. & Applic. 9: 91–108.MathSciNetzbMATHCrossRefGoogle Scholar
  8. Hannan, E. J. 1976. The identification and parameterization of ARMAX and state-space forms. Econometrics 44: 713–723.MathSciNetzbMATHCrossRefGoogle Scholar
  9. Harvey, A. C. 1981. The kalman filter and its applications in econometrics and time series analysis. In Methods of Operations Research, 44, [ed.] R. Henn, et al., Cambridge: Oelgeschlager, Gunn and Hain.Google Scholar
  10. Harvey, A. C. & G. D. A. Phillips. 1979. Maximum likelihood estimation of regression models with autoregressive moving average disturbances. Biometrika 66: 49–58.MathSciNetzbMATHGoogle Scholar
  11. Harvey, A. C. & G. D. A. Phillips. 1981. The estimation of regression models with time-varying parameters. In M. Deistier, E. F and G. Schwodiauer, [eds.], Games, Economic Dynamics, and Time Series Analysis, Physica-Verlag, Wein, pp. 306–321.Google Scholar
  12. Harvey, A. C. & R. G. Pierse. 1984. Estimating missing observations in economic time series. J. Amer. Statis Assoc. 79: 125–131.CrossRefGoogle Scholar
  13. Haugh, L. D. 1984. An introductory overview of some recent approaches to modeling spatial time series. In Time Series Analysis: Theory and Practice, [ed.] O. D. Anderson, New York: North-Holland.Google Scholar
  14. Jones, R. H. 1980. Maximum likelihood fitting of ARMA models to time series with missing observations. Technometrics 22: 389–395.MathSciNetzbMATHCrossRefGoogle Scholar
  15. Kennedy, W. J., Jr. & J. E. Gentle. 1980. Statistical Computing. New York: Marcel Dekker.Google Scholar
  16. Kohn, R. & C. F. Ansley. 1986. Estimation, prediction, and interpolation for ARMA models with missing data. J. Amer. Statist. Assoc. 81: 751–761.MathSciNetzbMATHCrossRefGoogle Scholar
  17. Liu, L. M. & D. M. Hanssens. 1981. A bayesian approach to time-varying cross-sectional regression models. J. Econom. 15: 341–356.MathSciNetzbMATHCrossRefGoogle Scholar
  18. Pfeifer, P. E. & S. J. Deutsch. 1980. A three-stage iterative procedure for space-time modeling. Technometrics 22: 35–47.zbMATHCrossRefGoogle Scholar
  19. Pfeifer, P. E. & S. J. Deutsch. 1980b. Identification and interpretation of first-order space-time ARMA models. Technometrics 22: 397–408.zbMATHCrossRefGoogle Scholar
  20. Shea, B. L. 1984. Maximum likelihood estimation of multivariate ARMA processes via the kalman filter. In Time Series Analysis: Theory and Practice, [ed.] O. D. Anderson, New York: North-Holland.Google Scholar
  21. Stoffer, D. S. 1986. Estimation and identification of space-time ARMAX models in the presence of missing data. J. Amer. Statist Assoc. 81: 762–772.MathSciNetzbMATHCrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1989

Authors and Affiliations

  • David K. Blough
    • 1
  1. 1.Department of Agricultural EconomicsUniversity of ArizonaTucsonUSA

Personalised recommendations