Data Analysis with Microsoft Excel: Updated for Office 2007

Chapter 11 Times Series 459

CONCEPT TUTORIALS

Smoothing Three-Parameter Exponential

The Exponential Smoothing workbook that you used earlier also contains

an interactive tutorial on two-parameter exponential smoothing.

To view the Exponential Smoothing workbook:

1 Return to the Exponential Smoothing file in the Explore folder.

Enable the macros in the workbook.

2 Scroll through the workbook until you reach “Explore Two-Parameter

Exponential Smoothing.” See Figure 11-17.

Figure 11-17
Exploring
two-
parameter
exponential
smoothing

forecasted values

relative weights

used in forecasting

the trend parameter

t

w

The worksheet shows data from a sample time series. The smoothing

factor for the location is equal to 0.15, as is the smoothing factor for trend.

The area curves at the bottom of the chart indicate the relative weights as-

signed to previous values in calculating the fi nal forecast for the location

and trend parameters. For t and w equal to 0.15, the most prominent obser-

vations occur within a few units of the current value. Earlier observations

have too little effect to be visible on the chart. On the basis of the two-

parameter exponential smoothing estimates, the forecasted value of the time

series is projected to be about 58.295, increasing at a rate of 0.042 points per

unit of time.