CHAPTER SUMMARY A time series is a set of observations on a qualitative variable collected over a time. Business very often requires forecasting time series variables. Independent variables are not available to build a regression model of a time series variable. In time series analysis, we analyze the past behavior of a variable in order to predict its future behavior.
Moving averages and exponential smoothing are widely used for routing short term forecasting. By making projections from past data, these methods expect that the future will resemble the past. The major components used are base level, trend, cyclic fluctuations. The measures of forecast accuracy can be calculated by MSE (the mean squared error between forecast and actual), MAD (the mean absolute deviation between forecast and actual) and MAPE (the mean absolute percent error between forecast and actual). The MAD calculation and the MAPE calculation are similar, one is absolute and other is relative.
MAPE is usually reserved for comparison in which the magnitudes of two cases are different. Out of the two exponential smoothing procedure is sophisticated enough to permit representations of a linear trend cyclical factors in its calculation. Exponential smoothing procedures are adaptive. Implementing an exponential smoothing procedure requires that initial values be specified and a smoothing factor to be chosen. The smoothing factor should be chosen to trade off stability and responsiveness in an appropriate manner. Although excel contains a data analysis tool for calculating moving- average forecasts and exponentially smoothed forecast, the tools do not accommodate the most powerful version of exponential smoothing which includes trends and cyclical components. We can take the exponential smoothing model further by including a cyclical factor.
For cyclical effect there are two models used additive model and multiplicative model.