A Strategy to Complete a Time Series with Missing Observations

  • Robert B. Miller
  • Osvaldo Ferreiro
Part of the Lecture Notes in Statistics book series (LNS, volume 25)


The problem introduced in this paper originated from a water pollution study of the Menomenee River in Milwaukee, Wisconsin. The final goal of the study was to estimate the total pollutant load deposited in Lake Michigan, for a given season. However, the information on the concentration of pollutants was not complete. Concentration data were available for only 20% of the days in the study period. Complete data were available on two related series: river flow rate and water equivalent of precipitation. Those series were used along with the pollutant concentration data to create a model for pollutant behavior (for the different pollutants and years), to estimate the missing observations and to estimate total pollutant loading for several seasons. (See Miller, et al (1980).)


Time Series Conditional Expectation Time Series Model ARMA Model Stationary Time Series 
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. Brubacker, S. and Wilson, G. (1976). Interpolating time series with applications to the estimation of holiday effects on electricity demand. JRSSC, 25, 107–116.Google Scholar
  2. Clinger, W. and Vanness, H.W. (1976). On unequally spaced time points in time series. Annals of Statistics, 4, 736–745.MathSciNetMATHCrossRefGoogle Scholar
  3. Dempster, A., Laird, N., and Rubin, D. (1977). Maximum likelihood from incomplete data via the E-M algorithm. JRSSB, 39, 1–22.MathSciNetMATHGoogle Scholar
  4. Dunsmuir, W., and Hannan, E.J. (1976). Vector linear time series models. Advances of Applied Probability, 8, 339–364.MathSciNetMATHCrossRefGoogle Scholar
  5. Dunsmuir, W., and Robinson, P.M. (1979). Estimation of time series models in the presence of missing data. Technical Report No. 9, M.I.T. Statistics Group, Mathematics Department.Google Scholar
  6. Dunsmuir, W., and Robinson, P.M. (1981a). Asymptotic theory for time series containing missing and amplitude modulated observations”. Sankhya, Ser. A, 43.Google Scholar
  7. Dunsmuir, W., and Robinson, P.M. (1981b). Parametric estimators for stationary time series with missing observations. Advances of Applied Probability, 13, 129–146.MathSciNetMATHCrossRefGoogle Scholar
  8. Jones, R.H. (1962). Spectral analysis with regularly missed observations. Annals of Mathematical Statistics, 33, 455–461.MathSciNetMATHCrossRefGoogle Scholar
  9. Jones, R.H. (1980). Maximum likelihood fitting of ARMA models to time series with missing observations. Technometrics 22, 389–395.MathSciNetMATHGoogle Scholar
  10. Miller, R., Bell, W., Ferreiro, O., and Wang, Y (1980). Regional inferences based on water quality monitoring data. Water Resources Center, University of Wisconsin, Technical Report No. 80-05.Google Scholar
  11. Parzen, E. (1963). “On spectral analysis with missing observations and amplitude modulation. Sankhya, Ser. A, 25, 383–392.MathSciNetMATHGoogle Scholar
  12. Robinson, P.M. (1977). Estimation of a time series model from unequally spaced data. Stochastic Processes and Their Applications, 6, 9–24.MathSciNetMATHCrossRefGoogle Scholar
  13. — (1980). Continuous model fitting from discrete data. Directions in Time Series, ed. D.R. Brillinger and G.C. Tiao, Institute of Mathematical Statistics, 263-278.Google Scholar
  14. Scheinok, P.S. (1965). Spectral analysis with randomly missed observations: The Binomial Case. Annals of Mathematical Statistics, 36, 971–977.MathSciNetMATHCrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1984

Authors and Affiliations

  • Robert B. Miller
    • 1
  • Osvaldo Ferreiro
    • 1
  1. 1.University of WisconsinUSA

Personalised recommendations