Modelling and forecasting flight delay at Kuala Lumpur International Airport using hybrid arima-garch model
Flight delay has become a hot issue over the recent years since it is one of the common factors that can impact the airline companies in terms of financial cost. When a flight is delayed, it requires the consumption of extra fuels, labor and other necessary aspects in the airline production process...
Saved in:
| 主要作者: | |
|---|---|
| 格式: | Thesis |
| 語言: | English |
| 出版: |
2019
|
| 主題: | |
| 在線閱讀: | http://eprints.utm.my/id/eprint/102403/1/IlyaFarhanaZulkefleeMFS2019.pdf.pdf |
| 標簽: |
添加標簽
沒有標簽, 成為第一個標記此記錄!
|
| 總結: | Flight delay has become a hot issue over the recent years since it is one of the common factors that can impact the airline companies in terms of financial cost. When a flight is delayed, it requires the consumption of extra fuels, labor and other necessary aspects in the airline production process and this may lead to higher operating cost to the airlines. Thus, this study aims to develop the hybridization between Autoregressive Integrated Moving Average (ARIMA) models and Generalised Autoregressive Conditional Heteroscedasticity (GARCH) models to predict the flight delay at Kuala Lumpur International Airport (KLIA). The weekly average minutes flight delay data were obtained from Kuala Lumpur Air Traffic Control Centre (KL ATCC) Flight Information Regions (FIR) Subang which dated from 5th May 2014 until 2nd July 2018. The data are divided into two parts, which 80% of the data are used as in-sample data and the rest 20% are used as out-sample data. The in-sample data are those from 5th May 2014 until 28th August 2017 and out-sample data will be from 4th September 2017 until 2nd July 2018. The data are first analysed by using GARCH models and the performance of these models is compared with hybrid ARIMA-GARCH models. The results of this study revealed that hybrid ARIMA-GARCH model is the best method for modelling and forecasting flight delay compared to GARCH models as it has a smaller value of Akaike’s Information Criterion, Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE). |
|---|