Commodity price prediction using neural networks
Commodities and its prices play a large role in the economic policies of countries whether they are exporters or importer. The ability to forecast commodity prices is then an important factor in decision making. The neural network is theorised to be able to trend non-linear and non-stationary time-s...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project |
| Language: | English |
| Published: |
2019
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/78064 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | Commodities and its prices play a large role in the economic policies of countries whether they are exporters or importer. The ability to forecast commodity prices is then an important factor in decision making. The neural network is theorised to be able to trend non-linear and non-stationary time-series data. Hence, this paper will evaluate the use of novel neural networks proposed by other researchers. The primary neural network examined are the recurrent neural networks. The use of EEMD (Ensemble Empirical Mode Decomposition) in neural networks were found to be generally positive. This paper then proposes using deeper networks architectures to further improve the use of EEMD in neural networks. |
|---|