Distribution-free consistency of kernel non-parametric M-estimators

Andrzej S. Kozek; Mirosław Pawlak

doi:10.1016/S0167-7152(02)00106-2

Distribution-free consistency of kernel non-parametric M-estimators

Andrzej S. Kozek^*, Mirosław Pawlak

^*Corresponding author for this work

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

3 Citations (Scopus)

Abstract

We prove that in the case of independent and identically distributed random vectors (X_i, Y_i) a class of kernel type M-estimators is universally and strongly consistent for conditional M-functionals. The term universal means that the strong consistency holds for all joint probability distributions of (X, Y). The conditional M-functional minimizes (2.2) for almost every x. In the case M (y) = y the conditional M-functional co-incides with the L₁-functional and with the conditional median.

Original language	English
Pages (from-to)	343-353
Number of pages	11
Journal	Statistics and Probability Letters
Volume	58
Issue number	4
DOIs	https://doi.org/10.1016/S0167-7152(02)00106-2
Publication status	Published - 15 Jul 2002

Access to Document

10.1016/S0167-7152(02)00106-2

Cite this

@article{c458310c8fc447f4a625614d0b207f20,

title = "Distribution-free consistency of kernel non-parametric M-estimators",

abstract = "We prove that in the case of independent and identically distributed random vectors (Xi, Yi) a class of kernel type M-estimators is universally and strongly consistent for conditional M-functionals. The term universal means that the strong consistency holds for all joint probability distributions of (X, Y). The conditional M-functional minimizes (2.2) for almost every x. In the case M (y) = y the conditional M-functional co-incides with the L1-functional and with the conditional median.",

author = "Kozek, \{Andrzej S.\} and Miros{\l}aw Pawlak",

year = "2002",

month = jul,

day = "15",

doi = "10.1016/S0167-7152(02)00106-2",

language = "English",

volume = "58",

pages = "343--353",

journal = "Statistics and Probability Letters",

issn = "0167-7152",

publisher = "Elsevier",

number = "4",

}

TY - JOUR

T1 - Distribution-free consistency of kernel non-parametric M-estimators

AU - Kozek, Andrzej S.

AU - Pawlak, Mirosław

PY - 2002/7/15

Y1 - 2002/7/15

N2 - We prove that in the case of independent and identically distributed random vectors (Xi, Yi) a class of kernel type M-estimators is universally and strongly consistent for conditional M-functionals. The term universal means that the strong consistency holds for all joint probability distributions of (X, Y). The conditional M-functional minimizes (2.2) for almost every x. In the case M (y) = y the conditional M-functional co-incides with the L1-functional and with the conditional median.

AB - We prove that in the case of independent and identically distributed random vectors (Xi, Yi) a class of kernel type M-estimators is universally and strongly consistent for conditional M-functionals. The term universal means that the strong consistency holds for all joint probability distributions of (X, Y). The conditional M-functional minimizes (2.2) for almost every x. In the case M (y) = y the conditional M-functional co-incides with the L1-functional and with the conditional median.

UR - https://www.scopus.com/pages/publications/0037100715

U2 - 10.1016/S0167-7152(02)00106-2

DO - 10.1016/S0167-7152(02)00106-2

M3 - Article

AN - SCOPUS:0037100715

SN - 0167-7152

VL - 58

SP - 343

EP - 353

JO - Statistics and Probability Letters

JF - Statistics and Probability Letters

IS - 4

ER -

Distribution-free consistency of kernel non-parametric M-estimators

Abstract

Access to Document

Other files and links

Fingerprint

Cite this