Wedding the wavelet transform and multivariate data analysis

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

We discuss the use of orthogonal wavelet transforms in preprocessing multivariate data for subsequent analysis, e.g., by clustering or dimensionality reduction. Wavelet transforms allow us to introduce multiresolution approximation, and multiscale nonparametric regression or smoothing, in a natural and integrated way into the data analysis. As will be explained in the first part of the paper, this approach is of greatest interest for multivariate data analysis when we use (i) datasets with ordered variables, e.g., time series, and (ii) object dimensionalities which are not too small, e.g., 16 and upwards. In the second part of the paper, a different type of wavelet decomposition is used. Applications illustrate the powerfulness of this new perspective on data analysis.

Original languageEnglish
Pages (from-to)161-183
Number of pages23
JournalJournal of Classification
Volume15
Issue number2
DOIs
Publication statusPublished - 1 Feb 1998
Externally publishedYes

Fingerprint

Multivariate Data Analysis
Wavelet Analysis
wedding
multivariate analysis
Wavelet Transform
Data analysis
data analysis
Multivariate Analysis
Nonparametric Smoothing
Wavelet Decomposition
Multivariate Data
Dimensionality Reduction
Nonparametric Regression
Multiresolution
Dimensionality
Preprocessing
Time series
Clustering
time series
Approximation

Cite this

@article{ce6f754a27554181b7b084af248730f7,
title = "Wedding the wavelet transform and multivariate data analysis",
abstract = "We discuss the use of orthogonal wavelet transforms in preprocessing multivariate data for subsequent analysis, e.g., by clustering or dimensionality reduction. Wavelet transforms allow us to introduce multiresolution approximation, and multiscale nonparametric regression or smoothing, in a natural and integrated way into the data analysis. As will be explained in the first part of the paper, this approach is of greatest interest for multivariate data analysis when we use (i) datasets with ordered variables, e.g., time series, and (ii) object dimensionalities which are not too small, e.g., 16 and upwards. In the second part of the paper, a different type of wavelet decomposition is used. Applications illustrate the powerfulness of this new perspective on data analysis.",
keywords = "Additive decomposition, Clustering, Data display, Dimensionality reduction, Exploratory data analysis, Partitioning, Preliminary data transformation, Principal components analysis, Wavelet transform",
author = "Fionn Murtagh",
year = "1998",
month = "2",
day = "1",
doi = "10.1007/s003579900029",
language = "English",
volume = "15",
pages = "161--183",
journal = "Journal of Classification",
issn = "0176-4268",
publisher = "Springer New York",
number = "2",

}

Wedding the wavelet transform and multivariate data analysis. / Murtagh, Fionn.

In: Journal of Classification, Vol. 15, No. 2, 01.02.1998, p. 161-183.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Wedding the wavelet transform and multivariate data analysis

AU - Murtagh, Fionn

PY - 1998/2/1

Y1 - 1998/2/1

N2 - We discuss the use of orthogonal wavelet transforms in preprocessing multivariate data for subsequent analysis, e.g., by clustering or dimensionality reduction. Wavelet transforms allow us to introduce multiresolution approximation, and multiscale nonparametric regression or smoothing, in a natural and integrated way into the data analysis. As will be explained in the first part of the paper, this approach is of greatest interest for multivariate data analysis when we use (i) datasets with ordered variables, e.g., time series, and (ii) object dimensionalities which are not too small, e.g., 16 and upwards. In the second part of the paper, a different type of wavelet decomposition is used. Applications illustrate the powerfulness of this new perspective on data analysis.

AB - We discuss the use of orthogonal wavelet transforms in preprocessing multivariate data for subsequent analysis, e.g., by clustering or dimensionality reduction. Wavelet transforms allow us to introduce multiresolution approximation, and multiscale nonparametric regression or smoothing, in a natural and integrated way into the data analysis. As will be explained in the first part of the paper, this approach is of greatest interest for multivariate data analysis when we use (i) datasets with ordered variables, e.g., time series, and (ii) object dimensionalities which are not too small, e.g., 16 and upwards. In the second part of the paper, a different type of wavelet decomposition is used. Applications illustrate the powerfulness of this new perspective on data analysis.

KW - Additive decomposition

KW - Clustering

KW - Data display

KW - Dimensionality reduction

KW - Exploratory data analysis

KW - Partitioning

KW - Preliminary data transformation

KW - Principal components analysis

KW - Wavelet transform

UR - http://www.scopus.com/inward/record.url?scp=0032264874&partnerID=8YFLogxK

U2 - 10.1007/s003579900029

DO - 10.1007/s003579900029

M3 - Article

VL - 15

SP - 161

EP - 183

JO - Journal of Classification

JF - Journal of Classification

SN - 0176-4268

IS - 2

ER -