Abstract
This paper examines three neural network models for forecasting to better understand the performance of neural networks for the case when the data exhibit a long memory pattern. The first model is simple one consisting of the time series values. The second model uses the same inputs as the first model with one additional input representing the combination (average) of five univariate time series forecasting models. The third model consists of all these individual time series forecasts in which the neural network is allowed to combine them on its own way. To obtain the optimum networks, the effect of network characteristics such as the training parameters, the number of hidden layers, and the testing and training percentages are simulated.
Original language | English |
---|---|
Pages | 359-364 |
Number of pages | 6 |
Publication status | Published - 1996 |
Event | Proceedings of the 1996 IEEE International Conference on Neural Networks, ICNN. Part 1 (of 4) - Washington, DC, USA Duration: Jun 3 1996 → Jun 6 1996 |
Other
Other | Proceedings of the 1996 IEEE International Conference on Neural Networks, ICNN. Part 1 (of 4) |
---|---|
City | Washington, DC, USA |
Period | 6/3/96 → 6/6/96 |
ASJC Scopus subject areas
- Software