Estimator

Also known as: Point estimator, Estimating function

An estimator $W=w(X_1,\ldots,X_n)$ is a rule — a function of the random sample — for guessing a parameter $\theta$. It is a random variable. The numerical value $w(x)$ on observed data is an estimate.

Same hat notation $\hat\theta$ is often used for both; context decides. Construct estimators via method of moments or MLE; evaluate via bias, MSE, and consistency.

← Back to CS Glossary