OATAO - Open Archive Toulouse Archive Ouverte Open Access Week

Prediction of U.S. General Aviation fatalities from extreme value approach

Diamoutene, Abdoulaye and Kamsu-Foguem, Bernard and Noureddine, Farid and Barro, Diakarya Prediction of U.S. General Aviation fatalities from extreme value approach. (2018) Transportation Research Part A: Policy and Practice, 109. 65-75. ISSN 0965-8564

(Document in English)

PDF (Author's version) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader

Official URL: https://doi.org/10.1016/j.tra.2018.01.022


General Aviation is the main component of the United States civil aviation and the most aviation accidents concern this aviation category. Between early 2015 and May 17, 2016, a total of 1546 general aviation accidents in the United States has left 466 fatalities and 384 injured. Hence, in this study, we investigate the risk of U.S. General Aviation accidents by examining historical U.S. General Aviation accidents. Using the Peak Over Threshold approach and Generalized Pareto Distribution, we predict the number of fatalities resulting in extreme GA accidents in the future operations. We use a graphical method and intensive parameters estimates to obtain the optimal range of the threshold. In order to assess the uncertainty in the inference and the accuracy of the results, we use the nonparametric bootstrap approach.

Item Type:Article
HAL Id:hal-02135111
Audience (journal):International peer-reviewed journal
Uncontrolled Keywords:
Institution:Université de Toulouse > Institut National Polytechnique de Toulouse - Toulouse INP (FRANCE)
Other partners > Institut polytechnique rural de formation et de recherche appliquée - IPR/IFRA (MALI)
Other partners > Université Ouaga 2 (BURKINA FASO)
Laboratory name:
Deposited On:02 Apr 2019 08:56

Repository Staff Only: item control page