Bayesian selection of primary resolution and wavelet basis functions for wavelet regression

Chun Gun Park, Hee Seok Oh, Hakbae Lee

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

This paper considers a Bayesian approach to selecting a primary resolution and wavelet basis functions. Most of papers on wavelet shrinkage have been focused on thresholding of wavelet coefficients, given a primary resolution which is usually determined by the sample size. However, it turns out that a proper primary resolution is much affected by the shape of an unknown function rather than by the sample size. In particular, Bayesian approaches to wavelet series suffer from computational burdens if the chosen primary resolution is too high. A surplus primary resolution may result in a poor estimate. In this paper, we propose a simple Bayesian method to determine a primary resolution and wavelet basis functions independently of the sample size. Results from a simulation study demonstrate the promising empirical properties of the proposed approach.

Original languageEnglish
Pages (from-to)291-302
Number of pages12
JournalComputational Statistics
Volume23
Issue number2
DOIs
Publication statusPublished - 2008 Apr 1

Fingerprint

Wavelet Bases
Basis Functions
Wavelets
Regression
Sample Size
Bayesian Approach
Wavelet Shrinkage
Wavelet Coefficients
Bayesian Methods
Thresholding
Simulation Study
Unknown
Series
Estimate
Demonstrate
Sample size

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Computational Mathematics

Cite this

@article{7bba1c1d01734948b9eac37318674b17,
title = "Bayesian selection of primary resolution and wavelet basis functions for wavelet regression",
abstract = "This paper considers a Bayesian approach to selecting a primary resolution and wavelet basis functions. Most of papers on wavelet shrinkage have been focused on thresholding of wavelet coefficients, given a primary resolution which is usually determined by the sample size. However, it turns out that a proper primary resolution is much affected by the shape of an unknown function rather than by the sample size. In particular, Bayesian approaches to wavelet series suffer from computational burdens if the chosen primary resolution is too high. A surplus primary resolution may result in a poor estimate. In this paper, we propose a simple Bayesian method to determine a primary resolution and wavelet basis functions independently of the sample size. Results from a simulation study demonstrate the promising empirical properties of the proposed approach.",
author = "Park, {Chun Gun} and Oh, {Hee Seok} and Hakbae Lee",
year = "2008",
month = "4",
day = "1",
doi = "10.1007/s00180-007-0055-y",
language = "English",
volume = "23",
pages = "291--302",
journal = "Computational Statistics",
issn = "0943-4062",
publisher = "Springer Verlag",
number = "2",

}

Bayesian selection of primary resolution and wavelet basis functions for wavelet regression. / Park, Chun Gun; Oh, Hee Seok; Lee, Hakbae.

In: Computational Statistics, Vol. 23, No. 2, 01.04.2008, p. 291-302.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Bayesian selection of primary resolution and wavelet basis functions for wavelet regression

AU - Park, Chun Gun

AU - Oh, Hee Seok

AU - Lee, Hakbae

PY - 2008/4/1

Y1 - 2008/4/1

N2 - This paper considers a Bayesian approach to selecting a primary resolution and wavelet basis functions. Most of papers on wavelet shrinkage have been focused on thresholding of wavelet coefficients, given a primary resolution which is usually determined by the sample size. However, it turns out that a proper primary resolution is much affected by the shape of an unknown function rather than by the sample size. In particular, Bayesian approaches to wavelet series suffer from computational burdens if the chosen primary resolution is too high. A surplus primary resolution may result in a poor estimate. In this paper, we propose a simple Bayesian method to determine a primary resolution and wavelet basis functions independently of the sample size. Results from a simulation study demonstrate the promising empirical properties of the proposed approach.

AB - This paper considers a Bayesian approach to selecting a primary resolution and wavelet basis functions. Most of papers on wavelet shrinkage have been focused on thresholding of wavelet coefficients, given a primary resolution which is usually determined by the sample size. However, it turns out that a proper primary resolution is much affected by the shape of an unknown function rather than by the sample size. In particular, Bayesian approaches to wavelet series suffer from computational burdens if the chosen primary resolution is too high. A surplus primary resolution may result in a poor estimate. In this paper, we propose a simple Bayesian method to determine a primary resolution and wavelet basis functions independently of the sample size. Results from a simulation study demonstrate the promising empirical properties of the proposed approach.

UR - http://www.scopus.com/inward/record.url?scp=42149099602&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=42149099602&partnerID=8YFLogxK

U2 - 10.1007/s00180-007-0055-y

DO - 10.1007/s00180-007-0055-y

M3 - Article

VL - 23

SP - 291

EP - 302

JO - Computational Statistics

JF - Computational Statistics

SN - 0943-4062

IS - 2

ER -