Iterative Method for Bandwidth Selection in Kernel Discriminant Analysis

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F60162694%3AG42__%2F14%3A00520841" target="_blank" >RIV/60162694:G42__/14:00520841 - isvavai.cz</a>
Result on the web
<a href="http://vavtest.unob.cz/registr" target="_blank" >http://vavtest.unob.cz/registr</a>
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
Iterative Method for Bandwidth Selection in Kernel Discriminant Analysis
Original language description
Kernel estimates belong to very effective nonparametric estimates of a probability density function. This concept provides an interesting alternative to the classical parametric approach in emipirical studies. Kernel estimates depend on a bandwidth matrix that controls smoothness of the estimated density. This paper is focused on the use of kernel estimates with the bandwidth matrix computed by the so-called 'iterative method' in kernel discriminant analysis. Its utility is illustrated through a short simulation study and real data applications, where standard model-based discrimination rules are compared with a kernel discriminant rule.
Czech name
—
Czech description
—

Project
—
Continuities
I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Publication year
2014
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection
32nd International Conference Mathematical Methods in Economics MME2014
ISBN
978-80-244-4209-9
ISSN
—
e-ISSN
—
Number of pages
6
Pages from-to
263-268
Publisher name
Palacký University, Olomouc
Place of publication
Olomouc
Event location
Olomouc
Event date
—
Type of event by nationality
EUR - Evropská akce
UT code for WoS article
000356417900046

Similar results(10)