Analysis and forecasting of temperature using time series forecasting methods a case study of Mus
The aim of this study is to forecast the daily average temperature of Mus province in Turkey using time series methods. The performance of three time series forecasting models is compared: LSTM, PROPHET, and ARIMA. The behavior of these models in temperature data is also investigated. It is found th...
Gespeichert in:
Veröffentlicht in: | Thermal science 2023, Vol.27 (4 Part B), p.3081-3088 |
---|---|
Hauptverfasser: | , |
Format: | Artikel |
Sprache: | eng |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | The aim of this study is to forecast the daily average temperature of Mus province in Turkey using time series methods. The performance of three time series forecasting models is compared: LSTM, PROPHET, and ARIMA. The behavior of these models in temperature data is also investigated. It is found that these methods give accurate results according to the MAE, MSE, and RMSE error metrics. However, LSTM produces slightly better results. The temperature data used in this study was obtained from the Mus Meteorology Provincial Directorate. Accurate temperature forecasting is important for many different areas, from energy, agriculture to water resource management. This study is an important research step in temperature analysis and forecasting, and it will contribute to relevant decision-making processes. |
---|---|
ISSN: | 0354-9836 2334-7163 |
DOI: | 10.2298/TSCI2304081T |