Forecasting food prices in Central Java with the hybrid ARIMAX calendar variation model
Variations in time series data that occur due to the effects of calendar factors such as holidays, seasons, or differences in the number of days in a month are referred to as calendar variations. Time series problems are not always linear or nonlinear. Instead, they sometimes contain both (linear...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Prince of Songkla University
2024-08-01
|
| Series: | Songklanakarin Journal of Science and Technology (SJST) |
| Subjects: | |
| Online Access: | https://sjst.psu.ac.th/journal/46-4/8.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Variations in time series data that occur due to the effects of calendar factors such as holidays, seasons, or differences in
the number of days in a month are referred to as calendar variations. Time series problems are not always linear or nonlinear.
Instead, they sometimes contain both (linear and nonlinear) at once hence hybrid models can be used to model time series problems.
Research was conducted on food prices in Central Java that are affected by calendar variations, namely beef and chicken egg price
data. The performance of these methods is compared using RMSE and MAPE accuracy measures. The hybrid ARIMAX-NN and
Reg-ARIMA methods produce the smallest RMSE and MAPE values compared to other methods, respectively.
|
|---|---|
| ISSN: | 0125-3395 |