Estimation of Variance of the Regression Estimator.

For estimating the variance of the regression estimator in simple random sampling without replacement, several design-based and model-based estimators and a new class of estimators are compared. Their second order expressions and biases are derived and compared. Empirical results on the biases and MSEs Mean Squared Errors of the variance estimators and the conditional and unconditional coverage probabilities of their associated t-intervals lend support to the theoretical results and suggest further questions. Originator-supplied keywords include Variance estimator, Design-based, Model-based, Jack Knife estimator, and Conditional coverage probabilities.