On prediction with time dependent arma models |
| |
| Authors: | M. Shelton Peiris |
| |
| Affiliation: | Department of Mathematics , Monash University , Clayton, Victoria, 3168, Australia |
| |
| Abstract: | The general nonstationary ARMA model with time dependent coefficients is considered and a necessary and sufficient condition for the existence of the solution of such a model is given. A simple form of the above solution is expressed as a one-sided-moving average, in terms of one-sided Green's functions. Using these results we solve the Prediction Problem and give some expressions for the h(>0) step-ahead linear predictor of such a process. A few examples are added to illustrate the theory. |
| |
| Keywords: | nonstationary time dependent predication green's function hilbert space innovation recursive relation time series |
|
|
