Time Series in M Dimensions: Spatial Models.
Abstract
The general theory of stationary spatial models is developed: namely MA, moving average; AR, autoregressive; and ARMA, autoregressive moving average processes. As compared to the time series in m dimensions, spatial models may be one-sided, two-sided, or mixed. Free use is made of the previous results of Aroian and his associates in time series in m dimensions. The main theoretical properties of the models in the univariate case are established. The multivariate case is even more important than the univariate. Estimation by minimum variance and simulation of the models are included. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 25, 1980
- Accession Number
- ADA091734
Entities
People
- Leo A. Aroian
- Omer Gebizlioglu
Organizations
- Union College