Time Series Forecasting of Energy Commodity using Grey Wolf Optimizer
The ability to model and perform decision making is an essential feature of many real-world applications including the forecasting of commodity prices. In this study, a forecasting model based on a relatively new Swarm Intelligence (SI) behaviour, namely Grey Wolf Optimizer (GWO), is developed for s...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English |
| Published: |
2015
|
| Subjects: | |
| Online Access: | http://umpir.ump.edu.my/id/eprint/8968/1/fskkp-2015-zuriani-time%20series%20forecasting.pdf http://umpir.ump.edu.my/id/eprint/8968/ http://www.iaeng.org/publication/IMECS2015/IMECS2015_pp25-30.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The ability to model and perform decision making is an essential feature of many real-world applications including the forecasting of commodity prices. In this study, a forecasting model based on a relatively new Swarm Intelligence (SI) behaviour, namely Grey Wolf Optimizer (GWO), is developed for short term time series forecasting. The model is built upon data obtained from the West Texas Intermediate (WTI) crude oil and gasoline price. Performance of the GWO model is compared against two other models which are developed based on Evolutionary Computation (EC) algorithms, namely the Artificial Bee Colony (ABC) and Differential Evolution (DE). Results showed that the GWO model outperformed DE in both crude oil and gasoline price forecasting. Furthermore, the proposed GWO produces a better forecast for gasoline price as compared to the ABC model,, as well as being at par in crude oil. Such an achievement indicates that GWO may become a competitor in the domain of time series forecasting and would be useful for investors in planning their investment and projecting their profit. |
|---|