A Time Series Analysis of Motor Vehicle Crash Fatalities in New York State

Abstract

Motor vehicle crashes are happening almost every day and unfortunately, some of them occur with fatality injuries. This paper presents a time series analysis on the trend and pattern of motor vehicle crash fatalities by month, day and day of the week in New York State. In this analysis, a statistical methodology for the decomposition of time series is used. The Kolmogorov – Zurbenko filter is used for decomposition of the crash fatalities into the long and seasonal and short-term component. Long-term component showed a 2 % decreases trend annually and seasonal term suggested increases of crash fatalities during the summer and decreases during winter period. The results indicate that sudden changes on trend are related to recessions that have occurred in USA. Another risk factor for crash fatalities in New York may be due to day light savings time that reduces light and visibility for drivers.