See also: Hypothesis Testing
Definitions:
- parameter :: numerical descriptive measure of a population. (usually unknown b/c populations are big)
- sample statistic :: numerical descriptive measure of a sample (like variance, stddev, etc)
- probability distribution :: the probability of every event in a population
- sampling distribution :: get statistics for a bunch of samples and model the statistic as a probability distribution
- point estimator :: rule/formula which says how to use sample data to to estimate a population parameter
- biased/unbiased estimates :: if a sample statistic is equal to the population parameter, it’s unbiased. Otherwise, it’s biased.
Population parameters vs sample statistics
| population param | sample stat | |
|---|---|---|
| mean | ||
| median | ||
| variance | ||
| standard deviation | ||
| binomial proportion |