|
|
 |
| |
|
|
|
On the Performance of the Mixed Seasonal Autoregressive Integrated Moving Average One-Dimensional Bilinear Time series model |
|
|
|
PP: 127-140 |
|
|
|
doi:10.18576/jsapl/070303
|
|
|
|
Author(s) |
|
|
|
Lanrewaju O. Adekola,
Johnson F. Ojo,
|
|
|
|
Abstract |
|
|
| The seasonal bilinear time series model is an effective tool in statistical analysis of seasonal time series. However, the existing pure Seasonal Autoregressive Integrated Moving Average (SARIMA) bilinear model focused only on series at the peak of seasons, excluding the pre-peak and post-peak of the seasons. This results in loss of vital information about the historical behaviour of the series. Hence, we examine the performance of a nonlinear model which is capable of explaining the behaviour of a seasonal series pre and post-peak of seasons. Nonlinear least-squares method of minimizing errors and Newton-Raphson iterative procedure were employed in estimating its parameters. Monthly rainfall series from the Nigerian Meteorological Agency between 1984-2016 was used to investigate the series at length of seasons, s= 1, 2, 3, 4, 6 and 12. Monte-Carlo simulation procedure was employed for sample sizes, n=250, 500 and 1000, each replicated 50 times. Forecast values were compared with the original series using the Mean Absolute Error (MAE). Akaike and Bayesian Information Criteria (AIC and BIC) were used to determine the optimum model order. The Mixed SARIMA One-Dimensional Bilinear (MSARIMAODBL) model performed excellently in explaining the behaviour of a nonlinear seasonal series before, at and after the peak of seasons.
|
|
|
|
|
|
|
|
