The transfer function models have an important role in multivariate time series analysis and it consider as one of the subjects that have an urgent application ,in the economic, social, industrial and other fields.
This study was applied on monthly data for the period (2011 to 2015) for time series representing death accidents and traffic damage using the descriptive approach to the realization of the hypotheses, we found that the two-time time series of traffic accidents are stationary, and we concluded that the estimated model of the conversion function is
y_t=(4.642-.208β)/(1+.419β+.986β^2@) β^2 xt+(1+2.216β+1.418β^2)/(1+1.8β)
Based on the criteria (RMSE, MAPE MAE, Normalized BIC), On the model, the traffic accidents occurring in the main roads were calculated.
