Method and model comparison by sum of ranking differences in cases of repeated observations (ties)

Klára Kollár-Hunek, K. Heberger

Research output: Article

71 Citations (Scopus)

Abstract

Significant progress has been achieved since the introduction of the new similarity measure: the sum of absolute ranking differences (SRDs) [TrAC - Trends in Anal. Chem. 29 (2010) 101-109]. Empirical evidences were accumulated about scaling, selection of the reference (benchmark) vector, cross-validation and grouping of variables (features, models, methods, etc.). The theory has been developed including the repeated observations (ties):. (i)The exact theoretical distribution (null distribution) for 48) an approximation has been developed using the Gaussian distribution fitted on the SRD distribution given by generating of three million n-dimensional random vectors.The validity and features of the SRD methodology with ties are illustrated using two case studies: evaluation of a sensory panel and ranking of financial indicators.

Original languageEnglish
Pages (from-to)139-146
Number of pages8
JournalChemometrics and Intelligent Laboratory Systems
Volume127
DOIs
Publication statusPublished - 2013

Fingerprint

Gaussian distribution

ASJC Scopus subject areas

  • Analytical Chemistry
  • Computer Science Applications
  • Software
  • Process Chemistry and Technology
  • Spectroscopy

Cite this

@article{359b620689894359b32f4fc49735d7bc,
title = "Method and model comparison by sum of ranking differences in cases of repeated observations (ties)",
abstract = "Significant progress has been achieved since the introduction of the new similarity measure: the sum of absolute ranking differences (SRDs) [TrAC - Trends in Anal. Chem. 29 (2010) 101-109]. Empirical evidences were accumulated about scaling, selection of the reference (benchmark) vector, cross-validation and grouping of variables (features, models, methods, etc.). The theory has been developed including the repeated observations (ties):. (i)The exact theoretical distribution (null distribution) for 48) an approximation has been developed using the Gaussian distribution fitted on the SRD distribution given by generating of three million n-dimensional random vectors.The validity and features of the SRD methodology with ties are illustrated using two case studies: evaluation of a sensory panel and ranking of financial indicators.",
keywords = "Data pre-treatment, Feature selection, Method comparison, Permutation test, Ranking, Sensory evaluation",
author = "Kl{\'a}ra Koll{\'a}r-Hunek and K. Heberger",
year = "2013",
doi = "10.1016/j.chemolab.2013.06.007",
language = "English",
volume = "127",
pages = "139--146",
journal = "Chemometrics and Intelligent Laboratory Systems",
issn = "0169-7439",
publisher = "Elsevier",

}

TY - JOUR

T1 - Method and model comparison by sum of ranking differences in cases of repeated observations (ties)

AU - Kollár-Hunek, Klára

AU - Heberger, K.

PY - 2013

Y1 - 2013

N2 - Significant progress has been achieved since the introduction of the new similarity measure: the sum of absolute ranking differences (SRDs) [TrAC - Trends in Anal. Chem. 29 (2010) 101-109]. Empirical evidences were accumulated about scaling, selection of the reference (benchmark) vector, cross-validation and grouping of variables (features, models, methods, etc.). The theory has been developed including the repeated observations (ties):. (i)The exact theoretical distribution (null distribution) for 48) an approximation has been developed using the Gaussian distribution fitted on the SRD distribution given by generating of three million n-dimensional random vectors.The validity and features of the SRD methodology with ties are illustrated using two case studies: evaluation of a sensory panel and ranking of financial indicators.

AB - Significant progress has been achieved since the introduction of the new similarity measure: the sum of absolute ranking differences (SRDs) [TrAC - Trends in Anal. Chem. 29 (2010) 101-109]. Empirical evidences were accumulated about scaling, selection of the reference (benchmark) vector, cross-validation and grouping of variables (features, models, methods, etc.). The theory has been developed including the repeated observations (ties):. (i)The exact theoretical distribution (null distribution) for 48) an approximation has been developed using the Gaussian distribution fitted on the SRD distribution given by generating of three million n-dimensional random vectors.The validity and features of the SRD methodology with ties are illustrated using two case studies: evaluation of a sensory panel and ranking of financial indicators.

KW - Data pre-treatment

KW - Feature selection

KW - Method comparison

KW - Permutation test

KW - Ranking

KW - Sensory evaluation

UR - http://www.scopus.com/inward/record.url?scp=84880681888&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84880681888&partnerID=8YFLogxK

U2 - 10.1016/j.chemolab.2013.06.007

DO - 10.1016/j.chemolab.2013.06.007

M3 - Article

AN - SCOPUS:84880681888

VL - 127

SP - 139

EP - 146

JO - Chemometrics and Intelligent Laboratory Systems

JF - Chemometrics and Intelligent Laboratory Systems

SN - 0169-7439

ER -