Forecasting

Simple Moving Average

Moving averages are often useful in recording of forecasting sales figures, expenses or manufacturing volume. There are many different types of moving average calculations. An often used, straightforward method of calculation is presented here.

In a moving average a specified number of data points are averaged. When there is a new piece of input data, the oldest piece of data is discarded to make room for the latest input. This replacement scheme makes the moving average a valuable tool in following trends. The fewer the number of data points, the more trend sensitive the average becomes. With a large number of data points, the average behaves more like a regular average, responding slowly to new input data.

A simple moving average may be calculated with your HP 12C as follows.

1.Press CLEAR .

2.Key in the first m data points (where m is the number of data points in the

average) and press after each entry.

3.Press to obtain the first average.

4.Key in the oldest (first value) entered in step 2 and press .

5.Key in the newest data point (m + 1) and press .

6.Press to obtain the next value of the moving average.

7.Repeat steps 4 through 5 for the remaining data.

Example: An electronics sales firm wished to calculate a 3-month moving average for the dollar volume of components sole each month. Sales for the first six months of this year were:

January

$211,570

February

112,550

March

190,060

April

131,760

May

300,500

June

271,120

Keystrokes

Display

71

Page 72
Image 72
HP 12C manual Forecasting, Simple Moving Average