Overview |
|
Definitions |
|
Factors for Control charts. |
|
Moving average calculations. |
|
Exponentially weighted moving average chart calculations. |
Moving averages are generally used to smooth out the noise in a time series and forecast future values of the series.
A MA (Moving Average) chart can be drawn based on either:
Subgroup means or
Individual observations (Subgroup size =1).
A moving average is calculated by averaging consecutive groups of observations in a series.
Example
A series begins with the numbers 8,9,10,11, and 12 and the moving average length is 3.
The values of the moving average are as follows:
Value |
Is |
First |
Missing |
Second |
Missing |
Third |
8,9, and 10 |
Fourth |
9,10, and 11 |
Fifth |
10, 11, and 12 and so on. |
Terms used in moving average calculations are:
Subgroup Size |
A set of one or more observations that are used to calculate a value on an SPC chart. Any subgroup size between 1 and 10 can be selected on the Chart tab in the SPC Control Properties dialog box. |
Length of MA |
Length of the moving averages. The value that is specified will be the number of subgroup means that will be included in each average. Any Length between 1 and 24 can be selected on the Chart tab in the SPC Control Properties dialog box. |
Weighting Factor |
Weight to be used in the exponentially weighted moving average. The value specified must be a number between 0 and 1. |
Charts
Exponentially weighted moving average chart calculations. |
|
Moving average calculations. |
|
Moving range calculations. |
SPC chart calculations. |