Let's Put the Seasonality and Trend in Decomposition

When discussing forecasting, most textbooks in management science or operations management explain that the values in a time series are a function of seasonality, trend, and random variation. Many textbooks have pictures depicting different patterns of data. We present a workbook that allows the user to manipulate a dynamic picture of these patterns and, in so doing, allows the user to examine the relationship between error measures and forecasting techniques for these different patterns. We also show a result regarding the relationship between the annual trend and the quarterly trend that surprises most of our students.