Ajustamento da série de acidentes aéreos mundiais através dos modelos Holt-Winters

Abstract

This study presents a modeling of time series data of air accidents in the period from July 2000 until June 2013 data were provided by the Directorate of Statistics and Census Service (DSEC), and an analysis was applied by smoothing models exponential. For this technique of exponential smoothing models used the Holt-Winters additive models, in which the smoothing constants are determined to minimize the mean square error between predicted and observed values; this analysis allowed the identification of components such as seasonality and atypical periods. The adjusted models were suitable for plotting forecasts of variables. The large number of vehicles, ease, lower cost, comfort, travel time are factors that influence to use this means of transport, however, the index of TIENTS has become a considerable variable when it comes to air travel. Thus the use of statistical methods became necessary for people to really analyze and explain such data.