A mixed integer programming approach for accommodating outliers with the minimax absolute error criterion (Q1174617)
From MaRDI portal
scientific article
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | A mixed integer programming approach for accommodating outliers with the minimax absolute error criterion |
scientific article |
Statements
A mixed integer programming approach for accommodating outliers with the minimax absolute error criterion (English)
0 references
25 June 1992
0 references
The minimax absolute error (MMAE) criterion in regression has a suitable property of treating errors on the equal basis, while the ordinary least squares method weights larger errors disproportionately. However, the former is sensitive to influence of outliers. This paper discusses the mixed integer approach paired with the MMAE criterion to provide simultaneous determination of candidate outliers along with model parameter estimation in a minimax problem. A numerical example illustrates this approach.
0 references
mixed integer programming
0 references
minimax absolute error criterion
0 references
least squares method
0 references
curve fitting
0 references
data fitting
0 references
regression
0 references
outliers
0 references
parameter estimation
0 references
minimax problem
0 references
numerical example
0 references
