Time series analysis

Abstract

The main goal of applying time series models is to obtain projections for the future values of the variable being studied, The choice of the most suitable model depends on an analysis of the values observed in the series, The present chapter introduces the use of time series models known as "exponential smoothing models", Such models, as far as parameter estimation is conceded, are much simpler and easier to understand than other models. The chapter presents the methods for statistical modeling of static series (simple exponential smoothing), series with a linear trend (double exponential smoothing) and series with seasonal trend (multiplicative and additive seasonal series). A comparison between the models of time series and of linear regression is also presented.