A statistic whose distribution depends on the parameter $\theta$.

A statistic whose distribution does not depend on the parameter $\theta$.

A statistic that is a sufficient estimator for $\theta$.

A statistic that is always biased.