Definition:M-Estimator
Jump to navigation
Jump to search
Definition
A robust estimator which gives maximum likelihood estimation when it is optimum, and are not greatly influenced by departures from requiremens for optimality, is called an $m$-estimator.
Such an estimator reduces or eliminates the influence of outliers.
Also see
- Results about $m$-estimators can be found here.
Sources
- 1998: David Nelson: The Penguin Dictionary of Mathematics (2nd ed.) ... (previous) ... (next): $m$-estimator
- 2008: David Nelson: The Penguin Dictionary of Mathematics (4th ed.) ... (previous) ... (next): $m$-estimator