Estimation, Prediction, and Interpolation for ARIMA Models with Missing Data (Q3746732)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 3982348
Language Label Description Also known as
default for all languages
No label defined
    English
    Estimation, Prediction, and Interpolation for ARIMA Models with Missing Data
    scientific article; zbMATH DE number 3982348

      Statements

      0 references
      0 references
      1986
      0 references
      estimation
      0 references
      prediction
      0 references
      diffuse initial conditions
      0 references
      maximum likelihood
      0 references
      transformation invariant likelihood
      0 references
      non-stationary Gaussian autoregressive integrated moving average (ARIMA) time series
      0 references
      missing observations
      0 references
      marginal likelihood
      0 references
      modified Kalman filter
      0 references
      mean squared error
      0 references
      Estimation, Prediction, and Interpolation for ARIMA Models with Missing Data (English)
      0 references
      The paper investigates non-stationary Gaussian autoregressive integrated moving average (ARIMA) time series models with missing observations. The marginal likelihood is defined and computed efficiently for these models. The definition of the likelihood for a non-stationary time series model is based on the concept of a marginal likelihood. This definition of the likelihood is invariant under a large class of transformations and if there are no missing observations this definition coincides with the definition given by Box and Jenkins. The computation of the marginal likelihood is carried out by using the univariate version of the modified Kalman filter introduced earlier by the authors [J. Stat. Comput. Simulation 21, 135-169 (1985; Zbl 0595.62092)]. The paper shows how to predict and interpolate missing observations and obtain the mean squared error of the estimate.
      0 references

      Identifiers