Definition

The error or residual term in a designed experiment or regression analysis represents the difference between the observed value and the value predicted by the fitted model. The error variance is the part of the process variance not accounted for by the Factors included in the model. It is also called unexplained, random or common-cause variation and includes variation from uncontrolled Factors, variation between Experimental Units and noise (environmental, temporal, etc.) variation.

Examples

 

See Also

Residual