Bandwidth matrix selectors for multivariate kernel density estimation

HASILOVÁ, Kamila, Ivanka HOROVÁ and Jan KOLÁČEK. Bandwidth matrix selectors for multivariate kernel density estimation. In C H Skiadas. Theoretical and Applied Issues in Statistics and Demography. Athens: International Society for the Advancement of Science and Technology (ISAST), 2014, p. 33-46. ISBN 978-618-81257-7-3.

Basic information

Original name

Bandwidth matrix selectors for multivariate kernel density estimation

Authors

HASILOVÁ, Kamila (203 Czech Republic, guarantor), Ivanka HOROVÁ (203 Czech Republic, belonging to the institution) and Jan KOLÁČEK (203 Czech Republic, belonging to the institution)

Edition

Athens, Theoretical and Applied Issues in Statistics and Demography, p. 33-46, 14 pp. 2014

Publisher

International Society for the Advancement of Science and Technology (ISAST)

---

Other information

Language

English

Type of outcome

Kapitola resp. kapitoly v odborné knize

Field of Study

10101 Pure mathematics

Country of publisher

Greece

Confidentiality degree

není předmětem státního či obchodního tajemství

Publication form

printed version "print"

RIV identification code

RIV/00216224:14310/14:00074755

Organization unit

Faculty of Science

ISBN

978-618-81257-7-3

Keywords in English

kernel density estimation; iterative method; integrated square error

Tags

AKR, rivok

Tags

International impact, Reviewed

---

Abstract

V originále

Multivariate kernel density estimators have become a common tool for empirical studies. These estimators depend on the choice of a crucial parameter -- the bandwidth matrix. In this paper, we propose an extension of the iterative method for finding the bandwidth matrix not only for density estimation but also for estimating its derivatives. This procedure is illustrated by a simulation study and two real data sets.