No label defined (Q4258792)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 1336714
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | No label defined |
scientific article; zbMATH DE number 1336714 |
Statements
14 September 1999
0 references
mixtures with varying concentrations
0 references
weighted empirical measures
0 references
kernel density estimators
0 references
nearest neighbour classification
0 references
Bayesian classification
0 references
0.8053774833679199
0 references
A mixture model with varying concentrations is considered. It means that there are independent observations \(\xi_1,\dots,\xi_n\), such that \( \text{Pr}\{\xi_j\in A\}=\sum_{k=1}^M w^k(j)H_k(A) \), where \(H_k\) are distributions of mixture components (unknown) and \(w^k(j)\) (known) are probabilities to choose the \(k\)-th component at the \(j\)-th moment (concentrations). The problem is to define to which component the observation \(\xi\) belongs if apriori probabilities \(w^k\) are given. Weighted empirical measures constructed by \(\xi_j\) are used to define some modifications of the kernel and the nearest neighbour estimates for densities of \(H_k\). Asymptotic properties of such estimates are investigated. It is proved that the classificators based on these estimates are asymptotically Bayesian. For the entire collection see [Zbl 0896.00028].
0 references