Partial least squares and compositional data: problems and alternatives

John Hinkle; William Rayens

doi:10.1016/0169-7439(95)00062-3

Partial least squares and compositional data: problems and alternatives

John Hinkle, William Rayens

Dr. Bing Zhang Department of Statistics

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

32 Scopus citations

Abstract

It is still widely unknown in chemometrics that the statistical analysis of compositional data requires fundamentally different tools than a similar analysis of unconstrained data. This article examines the problems that potentially occur when one performs a partial least squares (PLS) analysis on compositional data and suggests logcontrast partial least squares (LCPLS) as an alternative.

Original language	English
Pages (from-to)	159-172
Number of pages	14
Journal	Chemometrics and Intelligent Laboratory Systems
Volume	30
Issue number	1
DOIs	https://doi.org/10.1016/0169-7439(95)00062-3
State	Published - Nov 1995

Bibliographical note

Funding Information:
During the course of this research Professor Rayens was supported by NSF grant ATM-9108177.

Keywords

Compositional data
Partial least squares

ASJC Scopus subject areas

Software
Analytical Chemistry
Process Chemistry and Technology
Spectroscopy
Computer Science Applications

Access to Document

10.1016/0169-7439(95)00062-3

Cite this

@article{a3ca2efdf0cc4e178510532bd28c8736,

title = "Partial least squares and compositional data: problems and alternatives",

abstract = "It is still widely unknown in chemometrics that the statistical analysis of compositional data requires fundamentally different tools than a similar analysis of unconstrained data. This article examines the problems that potentially occur when one performs a partial least squares (PLS) analysis on compositional data and suggests logcontrast partial least squares (LCPLS) as an alternative.",

keywords = "Compositional data, Partial least squares",

author = "John Hinkle and William Rayens",

note = "Funding Information: During the course of this research Professor Rayens was supported by NSF grant ATM-9108177.",

year = "1995",

month = nov,

doi = "10.1016/0169-7439(95)00062-3",

language = "English",

volume = "30",

pages = "159--172",

number = "1",

}

TY - JOUR

T1 - Partial least squares and compositional data

T2 - problems and alternatives

AU - Hinkle, John

AU - Rayens, William

N1 - Funding Information: During the course of this research Professor Rayens was supported by NSF grant ATM-9108177.

PY - 1995/11

Y1 - 1995/11

N2 - It is still widely unknown in chemometrics that the statistical analysis of compositional data requires fundamentally different tools than a similar analysis of unconstrained data. This article examines the problems that potentially occur when one performs a partial least squares (PLS) analysis on compositional data and suggests logcontrast partial least squares (LCPLS) as an alternative.

AB - It is still widely unknown in chemometrics that the statistical analysis of compositional data requires fundamentally different tools than a similar analysis of unconstrained data. This article examines the problems that potentially occur when one performs a partial least squares (PLS) analysis on compositional data and suggests logcontrast partial least squares (LCPLS) as an alternative.

KW - Compositional data

KW - Partial least squares

UR - http://www.scopus.com/inward/record.url?scp=0028858284&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0028858284&partnerID=8YFLogxK

U2 - 10.1016/0169-7439(95)00062-3

DO - 10.1016/0169-7439(95)00062-3

M3 - Article

AN - SCOPUS:0028858284

SN - 0169-7439

VL - 30

SP - 159

EP - 172

JO - Chemometrics and Intelligent Laboratory Systems

JF - Chemometrics and Intelligent Laboratory Systems

IS - 1

ER -

Partial least squares and compositional data: problems and alternatives

Abstract

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this