Advances and Applications in Statistics
Volume 64, Issue 2, Pages 143 - 163
(October 2020) http://dx.doi.org/10.17654/AS064020143 |
|
A COMBINED FORECASTING MODEL FOR PREDICTING THE NUMBER OF ROAD TRAFFIC ACCIDENT DEATHS IN THAILAND
Jeeraporn Thaithanan and Chantha Wongoutong
|
Abstract: One of the most serious problems in Thailand is road safety. Many lives are lost on the roads every year. From 2011 to 2019, 188,905 people were killed in road accidents in Thailand. Thus reduction in the number of accidents and planning based on future forecast data are required. The aim of this study is to identify an appropriate model for forecasting the number of road traffic accident deaths in Thailand. Three combined forecasting techniques: Simple Average, Bates/Granger, Standard Eigenvector and four forecasting techniques: Holt-Winters, Box-Jenkins, Decomposition, and Artificial Neural Network were employed to generate individual forecasts. The four individual forecasting techniques were combined by assigning weights to each forecasting method to achieve a set of different combined forecasts. These methods were applied to predict the number of road traffic accident deaths in Thailand during 2011-2019 and their forecasting performances were evaluated via a number of metrics. The results suggest that the combined Bates/Granger method outperformed the others in terms of the root-mean-squared error, the mean-absolute-percentage error (MAPE), symmetric MAPE, and Theil’s U, with an average improvement of 25-30%. The results obtained in this study also indicate that this combined model is a powerful alternative method for predicting the number of road traffic accident deaths in Thailand. |
Keywords and phrases: forecasting method, combined forecasting method, road traffic accident deaths.
|
|
Number of Downloads: 287 | Number of Views: 750 |
|