Practical use of the Box-Jenkins methodology for seasonal financial data prediction

Abstract

Many economic/financial processes exhibit some form of seasonality. The agricultural, construction, and travel sectors have obvious seasonal patterns resulting from their dependence on the weather. Similarly, the Christmas holiday season has a pronounced influence on the retail trade. In fact, seasonal variation of a series may account for the preponderance of its total variance. Forecasts that ignore important seasonal patterns will have a high variance. One of the possibilities to implement quality forecasts in the seasonal data is to use the Box-Jenkins methodology, which seems to be a useful tool for this purpose. The research study of this paper is devoted to application of ARIMA/SARIMA models to the seasonal financial data that are sensitive to the mean shifting while calculating the autocorrelation in the data. Results are compared with other common models with appropriate commentary.