An asymptotic optimality of the discriminant procedure based on a nearest neighbor density estimator (Q1210060)
From MaRDI portal
 | This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: An asymptotic optimality of the discriminant procedure based on a nearest neighbor density estimator |
scientific article; zbMATH DE number 169024
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | An asymptotic optimality of the discriminant procedure based on a nearest neighbor density estimator |
scientific article; zbMATH DE number 169024 |
Statements
An asymptotic optimality of the discriminant procedure based on a nearest neighbor density estimator (English)
0 references
16 May 1993
0 references
We consider the discriminant procedure based on a nearest neighbor density estimator. It is shown that the procedure is asymptotically optimal in the sense that under an appropriate condition its Expected Cost of Misclassification (ECM) converges to that one of the Bayes procedure. The new discriminant procedure may be more interesting for practical applications because it is more directly, and demands a weaker condition.
0 references
expected cost of misclassification
0 references
ECM
0 references
nearest neighbor density estimator
0 references
discriminant procedure
0 references
0.7844151258468628
0 references
0.7473200559616089
0 references
0.746366024017334
0 references
