A KALMAN FILTER APPROACH TO THE FORECASTING OF MONTHLY TIME SERIES AFFECTED BY Morris Festivals

Abstract
Many economic time series are affected by the moving dates of festivals. In this paper a moving festival effect is defined and incorporated into a dynamic linear model which specifies how the parameters of several unobservable components of a time series evolve stochastically in time. The merits of this approach in comparison to other approaches are discussed and demonstrated using empirical data of three selected time series.

This publication has 11 references indexed in Scilit: