|
Comparative Analysis of Time Series Forecasting Models for Predicting Tourist Arrivals in Chiang Mai |
|---|---|
| รหัสดีโอไอ | |
| Creator | Rujipan Kosarat |
| Title | Comparative Analysis of Time Series Forecasting Models for Predicting Tourist Arrivals in Chiang Mai |
| Contributor | Tewa Promnuchanont, Worakarn Jaidee |
| Publisher | KKU Science Journal |
| Publication Year | 2567 |
| Journal Title | KKU Science Journal |
| Journal Vol. | 52 |
| Journal No. | 3 |
| Page no. | 289 - 302 |
| Keyword | Forecasting, Time Series Data, Model Performance |
| URL Website | https://ph01.tci-thaijo.org/index.php/KKUSciJ/article/view/257074 |
| Website title | Thai Journal Online (ThaiJO) |
| ISSN | 3027-6667 |
| Abstract | The purpose of this study is to evaluate how well five time series forecasting models—ARIMA, LSTM, Prophet, XGBoost, and Random Forest—predict Chiang Mai's arrivals of tourists. The study utilized a dataset that comprised visitor counts from January 2020 to December 2023. We split the data into two sets: a training set from January 2020 to December 2022, and a test set from January 2023 to December 2023. We used the mean absolute error (MAE), root mean square error (RMSE) and mean absolute percentage (MAPE) to evaluate the models' effectiveness. The results indicate that the ARIMA model demonstrated the highest accuracy. The Comparative analysis indicate that the ARIMA model exhibits the lowest forecasting error metrics among the models evaluated (MAE = 8,325.33, RMSE = 11,462.63, and MAPE = 10.16). |
