Moment convergence in regularized estimation under multiple and mixed-rates asymptotics

Hiroki Masuda, Y. Shimizu

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

In M-estimation under standard asymptotics, the weak convergence combined with the polynomial type large deviation estimate of the associated statistical random field Yoshida (2011) provides us with not only the asymptotic distribution of the associated M-estimator but also the convergence of its moments, the latter playing an important role in theoretical statistics. In this paper, we study the above program for statistical random fields of multiple and also possibly mixedrates type in the sense of Radchenko (2008) where the associated statistical random fields may be nondifferentiable and may fail to be locally asymptotically quadratic. Consequently, a very strong mode of convergence of a wide range of regularized M-estimators is ensured.Our results are applied to regularized estimation of an ergodic diffusion observed at high frequency.

Original languageEnglish
Pages (from-to)81-110
Number of pages30
JournalMathematical Methods of Statistics
Volume26
Issue number2
DOIs
Publication statusPublished - Apr 1 2017

Fingerprint

Random Field
M-estimator
Moment
M-estimation
Weak Convergence
Large Deviations
Asymptotic distribution
Statistics
Polynomial
Estimate
Range of data
Random field
Standards
Large deviations
Polynomials
Weak convergence

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Moment convergence in regularized estimation under multiple and mixed-rates asymptotics. / Masuda, Hiroki; Shimizu, Y.

In: Mathematical Methods of Statistics, Vol. 26, No. 2, 01.04.2017, p. 81-110.

Research output: Contribution to journalArticle

@article{7e78bd4083a243ee84c44b110cc8a9c9,
title = "Moment convergence in regularized estimation under multiple and mixed-rates asymptotics",
abstract = "In M-estimation under standard asymptotics, the weak convergence combined with the polynomial type large deviation estimate of the associated statistical random field Yoshida (2011) provides us with not only the asymptotic distribution of the associated M-estimator but also the convergence of its moments, the latter playing an important role in theoretical statistics. In this paper, we study the above program for statistical random fields of multiple and also possibly mixedrates type in the sense of Radchenko (2008) where the associated statistical random fields may be nondifferentiable and may fail to be locally asymptotically quadratic. Consequently, a very strong mode of convergence of a wide range of regularized M-estimators is ensured.Our results are applied to regularized estimation of an ergodic diffusion observed at high frequency.",
author = "Hiroki Masuda and Y. Shimizu",
year = "2017",
month = "4",
day = "1",
doi = "10.3103/S1066530717020016",
language = "English",
volume = "26",
pages = "81--110",
journal = "Mathematical Methods of Statistics",
issn = "1066-5307",
publisher = "Allerton Press Inc.",
number = "2",

}

TY - JOUR

T1 - Moment convergence in regularized estimation under multiple and mixed-rates asymptotics

AU - Masuda, Hiroki

AU - Shimizu, Y.

PY - 2017/4/1

Y1 - 2017/4/1

N2 - In M-estimation under standard asymptotics, the weak convergence combined with the polynomial type large deviation estimate of the associated statistical random field Yoshida (2011) provides us with not only the asymptotic distribution of the associated M-estimator but also the convergence of its moments, the latter playing an important role in theoretical statistics. In this paper, we study the above program for statistical random fields of multiple and also possibly mixedrates type in the sense of Radchenko (2008) where the associated statistical random fields may be nondifferentiable and may fail to be locally asymptotically quadratic. Consequently, a very strong mode of convergence of a wide range of regularized M-estimators is ensured.Our results are applied to regularized estimation of an ergodic diffusion observed at high frequency.

AB - In M-estimation under standard asymptotics, the weak convergence combined with the polynomial type large deviation estimate of the associated statistical random field Yoshida (2011) provides us with not only the asymptotic distribution of the associated M-estimator but also the convergence of its moments, the latter playing an important role in theoretical statistics. In this paper, we study the above program for statistical random fields of multiple and also possibly mixedrates type in the sense of Radchenko (2008) where the associated statistical random fields may be nondifferentiable and may fail to be locally asymptotically quadratic. Consequently, a very strong mode of convergence of a wide range of regularized M-estimators is ensured.Our results are applied to regularized estimation of an ergodic diffusion observed at high frequency.

UR - http://www.scopus.com/inward/record.url?scp=85021680214&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85021680214&partnerID=8YFLogxK

U2 - 10.3103/S1066530717020016

DO - 10.3103/S1066530717020016

M3 - Article

AN - SCOPUS:85021680214

VL - 26

SP - 81

EP - 110

JO - Mathematical Methods of Statistics

JF - Mathematical Methods of Statistics

SN - 1066-5307

IS - 2

ER -