Introduction to Statistical Modeling with SAS/STAT Software: Estimating the Error Variance

Estimating the Error Variance

The least squares principle does not provide for a parameter estimator for $\text{[math]}$ . The usual approach is to use a method-of-moments estimator that is based on the sum of squared residuals. If the model is correct, then the mean square for error, defined to be $\text{[math]}$ divided by its degrees of freedom,

	$\text{[math]}$	$\text{[math]}$
	$\text{[math]}$	$\text{[math]}$

is an unbiased estimator of $\text{[math]}$ .

Previous Page
|
Next Page
|
Top of Page