In last week's Blog, I presented one of the equations found in the BOK in less complicated terms. My purpose was to make the equation easier to remember and apply in case you have to use it to answer a calculation question on the CTP exam. So, in keeping with this practice of reducing equations, where possible, to their simplest terms, here is a similar discussion of the Exponential Smoothing equation.
This equation is one of the Statistical Time Series Forecasting techniques that is discussed in the BOK in connection with Cash Forecasting and can be found in the following exam preparation resources:
- AFP Learning System, Treasury – Participant Guide, Page S5-18, Slide 45 (Note: The input data for the calculation is on Slide 44).
- ETM 3rd Ed. Pages 286 – 288.
The Exponential Smoothing equation as presented in the BOK can be restated as follows:
Next Period Forecast = RF + [alpha (RA - RF)]
RF is the most recent forecast value.
RA is the most recent actual value.
“alpha” is the smoothing constant.
Using input from slide 44 and applying the restated Exponential Smoothing equation the day 8 forecast, for example, can be calculated as follows:
|Next Period Forecast (day 8)
||$120,760 + [.40 ($129,000 - $120,760)]
||$120,760 + [.40 ($8,240)]
||$120,760 + $3,296
Finally, here’s a couple of additional points to remember about this equation:
Remember the general rule, “always work equations from the inside out” as I’ve shown in the day 8 calculation. Where, first I calculated the value within the parentheses, then calculated the value within the brackets and then added this result to the $120,760.
The smoothing constant “alpha”, which is a value somewhere between zero and one, is selected by the forecaster. So, if you get the Exponential Smoothing calculation on the exam, the “alpha” to be used in the calculation will be given to you as part of the input data.
-George Schilling, CTP