Nonparametric density estimation for stratified samples

Robert Breunig

doi:10.1016/j.spl.2008.01.099

Nonparametric density estimation for stratified samples

Robert Breunig^*

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

9 Citations (Scopus)

Abstract

We consider a weighted, nonparametric density estimator for stratified samples. We derive the optimal bandwidth using information on within-stratum variances and means. We provide a plug-in bandwidth when all strata are normally distributed. We show that the optimal sampling scheme is stratified sampling proportional to size, irrespective of the stratum-specific densities.

Original language	English
Pages (from-to)	2194-2200
Number of pages	7
Journal	Statistics and Probability Letters
Volume	78
Issue number	14
DOIs	https://doi.org/10.1016/j.spl.2008.01.099
Publication status	Published - 1 Oct 2008

Access to Document

10.1016/j.spl.2008.01.099

Cite this

@article{39737ef6f6d5480c9993ae3db45b3f84,

title = "Nonparametric density estimation for stratified samples",

abstract = "We consider a weighted, nonparametric density estimator for stratified samples. We derive the optimal bandwidth using information on within-stratum variances and means. We provide a plug-in bandwidth when all strata are normally distributed. We show that the optimal sampling scheme is stratified sampling proportional to size, irrespective of the stratum-specific densities.",

author = "Robert Breunig",

year = "2008",

month = oct,

day = "1",

doi = "10.1016/j.spl.2008.01.099",

language = "English",

volume = "78",

pages = "2194--2200",

journal = "Statistics and Probability Letters",

issn = "0167-7152",

publisher = "Elsevier B.V.",

number = "14",

}

TY - JOUR

T1 - Nonparametric density estimation for stratified samples

AU - Breunig, Robert

PY - 2008/10/1

Y1 - 2008/10/1

N2 - We consider a weighted, nonparametric density estimator for stratified samples. We derive the optimal bandwidth using information on within-stratum variances and means. We provide a plug-in bandwidth when all strata are normally distributed. We show that the optimal sampling scheme is stratified sampling proportional to size, irrespective of the stratum-specific densities.

AB - We consider a weighted, nonparametric density estimator for stratified samples. We derive the optimal bandwidth using information on within-stratum variances and means. We provide a plug-in bandwidth when all strata are normally distributed. We show that the optimal sampling scheme is stratified sampling proportional to size, irrespective of the stratum-specific densities.

UR - http://www.scopus.com/inward/record.url?scp=50649115745&partnerID=8YFLogxK

U2 - 10.1016/j.spl.2008.01.099

DO - 10.1016/j.spl.2008.01.099

M3 - Article

SN - 0167-7152

VL - 78

SP - 2194

EP - 2200

JO - Statistics and Probability Letters

JF - Statistics and Probability Letters

IS - 14

ER -

Nonparametric density estimation for stratified samples

Abstract

Access to Document

Other files and links

Fingerprint

Cite this