Mixtures of weighted distance-based models for ranking data

Hong Lee; Philip L.H. Yu

doi:10.1007/978-3-7908-2604-3-52

Mixtures of weighted distance-based models for ranking data

Hong Lee, Philip L.H. Yu

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

2 Citations (Scopus)

Abstract

Ranking data has applications in different fields of studies, like marketing, psychology and politics. Over the years, many models for ranking data have been developed. Among them, distance-based ranking models, which originate from the classical rank correlations, postulate that the probability of observing a ranking of items depends on the distance between the observed ranking and a modal ranking. The closer to the modal ranking, the higher the ranking probability is. However, such a model basically assumes a homogeneous population, and the single dispersion parameter may not be able to describe the data very well. To overcome the limitations, we consider new weighted distance measures which allow different weights for different ranks in formulating more flexible distancebased models. The mixtures of weighted distance-based models are also studied for analyzing heterogeneous data. Simulations results will be included, and we will apply the proposed methodology to analyze a real world ranking dataset.

Original language	English
Title of host publication	Proceedings of COMPSTAT 2010 - 19th International Conference on Computational Statistics, Keynote, Invited and Contributed Papers
Publisher	Springer Berlin
Pages	517-524
Number of pages	8
ISBN (Print)	9783790826036
DOIs	https://doi.org/10.1007/978-3-7908-2604-3-52
Publication status	Published - 1 Jan 2010
Externally published	Yes
Event	19th International Conference on Computational Statistics, COMPSTAT 2010 - Paris, France Duration: 22 Aug 2010 → 27 Aug 2010

Conference

Conference	19th International Conference on Computational Statistics, COMPSTAT 2010
Country/Territory	France
City	Paris
Period	22/08/10 → 27/08/10

Keywords

Distance-based model
Mixture model
Ranking data

ASJC Scopus subject areas

Statistics and Probability

More information

10.1007/978-3-7908-2604-3-52

Cite this

@inproceedings{3ee2bfcf856d4f29b4ee4fc5461f1efc,

title = "Mixtures of weighted distance-based models for ranking data",

abstract = "Ranking data has applications in different fields of studies, like marketing, psychology and politics. Over the years, many models for ranking data have been developed. Among them, distance-based ranking models, which originate from the classical rank correlations, postulate that the probability of observing a ranking of items depends on the distance between the observed ranking and a modal ranking. The closer to the modal ranking, the higher the ranking probability is. However, such a model basically assumes a homogeneous population, and the single dispersion parameter may not be able to describe the data very well. To overcome the limitations, we consider new weighted distance measures which allow different weights for different ranks in formulating more flexible distancebased models. The mixtures of weighted distance-based models are also studied for analyzing heterogeneous data. Simulations results will be included, and we will apply the proposed methodology to analyze a real world ranking dataset.",

keywords = "Distance-based model, Mixture model, Ranking data",

author = "Hong Lee and Yu, {Philip L.H.}",

year = "2010",

month = jan,

day = "1",

doi = "10.1007/978-3-7908-2604-3-52",

language = "English",

isbn = "9783790826036",

pages = "517--524",

booktitle = "Proceedings of COMPSTAT 2010 - 19th International Conference on Computational Statistics, Keynote, Invited and Contributed Papers",

publisher = "Springer Berlin",

address = "Germany",

note = "19th International Conference on Computational Statistics, COMPSTAT 2010 ; Conference date: 22-08-2010 Through 27-08-2010",

}

Lee, H & Yu, PLH 2010, Mixtures of weighted distance-based models for ranking data. in Proceedings of COMPSTAT 2010 - 19th International Conference on Computational Statistics, Keynote, Invited and Contributed Papers. Springer Berlin, pp. 517-524, 19th International Conference on Computational Statistics, COMPSTAT 2010, Paris, France, 22/08/10. https://doi.org/10.1007/978-3-7908-2604-3-52

Mixtures of weighted distance-based models for ranking data. / Lee, Hong; Yu, Philip L.H.
Proceedings of COMPSTAT 2010 - 19th International Conference on Computational Statistics, Keynote, Invited and Contributed Papers. Springer Berlin, 2010. p. 517-524.

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

TY - GEN

T1 - Mixtures of weighted distance-based models for ranking data

AU - Lee, Hong

AU - Yu, Philip L.H.

PY - 2010/1/1

Y1 - 2010/1/1

N2 - Ranking data has applications in different fields of studies, like marketing, psychology and politics. Over the years, many models for ranking data have been developed. Among them, distance-based ranking models, which originate from the classical rank correlations, postulate that the probability of observing a ranking of items depends on the distance between the observed ranking and a modal ranking. The closer to the modal ranking, the higher the ranking probability is. However, such a model basically assumes a homogeneous population, and the single dispersion parameter may not be able to describe the data very well. To overcome the limitations, we consider new weighted distance measures which allow different weights for different ranks in formulating more flexible distancebased models. The mixtures of weighted distance-based models are also studied for analyzing heterogeneous data. Simulations results will be included, and we will apply the proposed methodology to analyze a real world ranking dataset.

AB - Ranking data has applications in different fields of studies, like marketing, psychology and politics. Over the years, many models for ranking data have been developed. Among them, distance-based ranking models, which originate from the classical rank correlations, postulate that the probability of observing a ranking of items depends on the distance between the observed ranking and a modal ranking. The closer to the modal ranking, the higher the ranking probability is. However, such a model basically assumes a homogeneous population, and the single dispersion parameter may not be able to describe the data very well. To overcome the limitations, we consider new weighted distance measures which allow different weights for different ranks in formulating more flexible distancebased models. The mixtures of weighted distance-based models are also studied for analyzing heterogeneous data. Simulations results will be included, and we will apply the proposed methodology to analyze a real world ranking dataset.

KW - Distance-based model

KW - Mixture model

KW - Ranking data

UR - http://www.scopus.com/inward/record.url?scp=84904096013&partnerID=8YFLogxK

U2 - 10.1007/978-3-7908-2604-3-52

DO - 10.1007/978-3-7908-2604-3-52

M3 - Conference article published in proceeding or book

SN - 9783790826036

SP - 517

EP - 524

BT - Proceedings of COMPSTAT 2010 - 19th International Conference on Computational Statistics, Keynote, Invited and Contributed Papers

PB - Springer Berlin

T2 - 19th International Conference on Computational Statistics, COMPSTAT 2010

Y2 - 22 August 2010 through 27 August 2010

ER -

Mixtures of weighted distance-based models for ranking data

Abstract

Conference

Keywords

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this