Comparing the accuracy of different approaches of Holt Winters 'ARIMA ' exponential and intelligent in anticipation of accidents

Annual a large number of people lose their lives in the accident. In this paper, use death statistics of road accidents between the years 2009 - 2013 to compare the accuracy of forecasting methods Holt Winters multiplicative and cumulative, time series ARIMA, exponential adjusted and intelligent method of artificial neural network has been studied. Criteria for selecting the best forecasting method and the average percentage error have been selected. The results showed that ARIMA time series model more accurately forecast the death rate caused to accidents. . The forecasted values for the years 2014 and 2015 and 2016 show that the number of death caused accidents in the country with gentle slopes on the decline which can be taught and proper management of this number decreased by more slope