The place of the \(L_ 1\)-norm in robust estimation (Q1091694)
From MaRDI portal
scientific article
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | The place of the \(L_ 1\)-norm in robust estimation |
scientific article |
Statements
The place of the \(L_ 1\)-norm in robust estimation (English)
0 references
1987
0 references
In robust estimation, \(L_ 1\)-methods serve two main purposes: They provide estimates with minimal bias if the observations are asymmetrically contaminated, and they furnish convenient starting values for estimates based on iterative procedures. In general, such \(L_ 1\)- estimates are weighted ones, but there are problems with the choice of weights. It is pointed out that the two most popular choices of weights are non- robust: with constant weights, the estimates are sensitive to outliers at influential points, and with Hampel-Krasker-Welsch-type weights they are sensitive to local shifts. Other M-estimates of regression suffer from the same problem. Some recommendations for qualitative improvement are given.
0 references
regression diagnostics
0 references
breakdown point
0 references
sensitivity
0 references
generalization of sample median
0 references
\(L_ 1\)-estimation
0 references
minimax asymptotic bias
0 references
multiple regression
0 references
robust estimation
0 references
minimal bias
0 references
iterative procedures
0 references
choice of weights
0 references
constant weights
0 references
outliers
0 references
influential points
0 references
Hampel- Krasker-Welsch-type weights
0 references
local shifts
0 references
M-estimates
0 references
