Trend-Adjusted Exponential Smoothing

Trend-Adjusted Exponential Smoothing (TAES) is a forecasting technique that uses information regarding (1) average business volumes in the past, and (2) trends observed in volumes in the past (up-swing or down-turn) to develop a forecast of future demand. This method works as follows:

• It incorporates a weighted average of actual business volumes using up to 15 data points.
• It incorporates a weighted average of trends—up-swings and down-turns—in business volume.
• It employs mathematical logic to determine the optimal weights to be used for both average business volume and average trend that result in minimized forecast error over the time period analyzed. 

When applied, the TAES modeling algorithm analyzes the last reported 15 data points for Actual KBIs (that is, the prior 15 weeks of data) and calculates a trend factor for the upcoming week. If you want to forecast more than one week in the future, the TAES modeling algorithm extrapolates the trend-factor calculation across the desired forecast range.  

EXAMPLE

The TAES-calculated trend factor of 20 is applied to an 8-week forecast range.