Bias

From Displayr
Jump to: navigation, search

The difference between an estimator's expected value and its actual value. A statistic is said to be biased if it is calculated in such a way that its result can be expected to be systematically different from what it is trying to estimate. For example, if you attempted to estimate the future height of a child based on the average height of the child's parents, the estimate would be biased because, over time, generations get taller so this estimate is biased downwards.