Theorem
The sum of squared deviations is least about the mean
Statistics • Measures of Dispersion
Statement
For observations with mean , , so it is least exactly when .
How to use it
- 1Why it works: the term is never negative and is only at ; the other term does not depend on .
- 2If the expression is , take out : , so , that is .
Worked example
For which is least, if ?
Step-by-step
- 1
- 2Least when
Why: Squared deviations are least about the mean.
- 3
Answer
Watch out
The median makes least, not .
Builds on
Also known as
least squares property of mean, minimum of sum of squares of deviations, mean minimises squared deviation