Time Series Analysis and Its Applications to Data on Traf c Accidents
Abstract
The purpose of this study is to analyze the road traf c accidents in Turkey through time series and to predict the number of the prospective road traf c accidents by determining the most appropriate time series model. A time series was created with the data from the records of Turkish Statistical Institute related to the numbers of the road traf c accidents happened in Turkey between 1955 and 2012. It was seen in the study that the difference of the series itself and its rst difference from the autocorrelation function graph were not stable and the series be- came stable after the second difference was taken. Augmented Dickey-Fuller test was carried out for the stability test. In order to de ne the model appropriateness, whether the autocorrelation graph had white noise or not and the results of Box-Ljung test were taken into consideration. Model predictions were made from previously tested mod- els whose parameter predictions were signi cant and Akaike Information Criterion (AIC) and Schwartz Bayesian Information Criterion (BIC) values were the lowest. The most appropriate prediction model de ned for the road traf c accidents is the one called ARIMA (0, 2, 3) which is an integrated moving average model with a third de- gree mobility. In accordance with this model, it is predicted that the road traf c accidents in Turkey will increase consistently from 2013 to 2020 and the number in 2013 will be 1421791 and 2049307 in 2020.
Keywords
References
- Akdi, Y., 2010. Zaman Serileri Analizi (Birim Kökler ve Kointegrasyon). Gazi Kitabevi, Ankara, 27.
- Akkaya, Ş., Altıntaş, H., 2001. Türkiye’de karayolu tra k kazaları istatistik analizi: 1989-1999. V. Ulusal Ekonometri ve İstatistik Sempozyumu, 12-19 Eylül 2001, Adana.
- Anonim, 2013. Genel kaza istatistikleri. http://www.tra k.gov.tr/ Sayfalar/Istatistikler/Genel-Kaza.aspx
- Atalay, A., Tortum, A., Gökdağ, M., 2012. Türkiye’de 1977-2006 yılları arasında meydana gelen aylık tra k kazalarının zamansal analizi. Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi, 18(3):221-229.
- Bowerman, B.L., O’Connell, R.T., 1993. Forecasting and Time Series: An Applied Approach. Duxbury Press. Box, G.E. P., Jenkins, G. M., 1976. Time series analysis forecasting and control revised edition. San Francisco: Holden Day, p. 25-36.
- Brockwell, P.J., Davis, R.A., 2006. Time series: Theory and methods. Springer, New York, 78, 310-312.
- Brockwell, P. J., Davis, R.A., 1996. Introduction time series and forecasting. Springer Texts in Statistics, Springer Verlag New York Inc., 35.
- Çodur, M. Y., Tortum, A., Çodur, M., 2013. Genelleştirilmiş Lineer Regresyon ile Erzurum Kuzey Çevre Yolu Kaza Tahmin Modeli, Iğdır Üniversitesi Fen Bilimleri Enstitüsü Dergisi, 3(1): 79-84.
Details
Primary Language
English
Subjects
-
Journal Section
Research Article
Authors
Şenol Çelı̇k
*
Türkiye
Publication Date
December 31, 2013
Submission Date
May 21, 2013
Acceptance Date
August 25, 2013
Published in Issue
Year 2013 Volume: 3 Number: 4