Similarity-based model for ordered categorical data

Gabi Gayer; Offer Lieberman; Omer Yaffe

doi:10.1080/07474938.2017.1308054

Similarity-based model for ordered categorical data

Department of Economics

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

5 Scopus citations

Abstract

In a large variety of applications, the data for a variable we wish to explain are ordered and categorical. In this paper, we present a new similarity-based model for the scenario and investigate its properties. We establish that the process is ψ-mixing and strictly stationary and derive the explicit form of the autocorrelation function in some special cases. Consistency and asymptotic normality of the maximum likelihood estimator of the model’s parameters are proven. A simulation study supports our findings. The results are applied to the Netflix data set, comprised of a survey on users’ grading of movies.

Original language	English
Pages (from-to)	263-278
Number of pages	16
Journal	Econometric Reviews
Volume	38
Issue number	3
DOIs	https://doi.org/10.1080/07474938.2017.1308054
State	Published - 16 Mar 2019

Bibliographical note

Publisher Copyright:
© 2017, © 2017 Taylor & Francis Group, LLC.

Funding

The author gratefully acknowledges the support from Israel Science Foundation grant No. 1188/14. The author gratefully acknowledges the support from Israel Science Foundation grant No. 1082/14 and from the Sapir Center in Tel Aviv University. The author gratefully acknowledges the support from Israel Science Foundation grant No. 1188/14. The author gratefully acknowledges the support from Israel Science Foundation grant No. 1082/14 and from the Sapir Center in Tel Aviv University.

Funders	Funder number
Israel Science Foundation	1188/14, 1082/14
Center for Nanoscience and Nanotechnology
Cancer Biology Research Center, Tel Aviv University

Keywords

Consistency
ergodicity
mixing
ordered probit
similarity
stationarity

Access to Document

10.1080/07474938.2017.1308054

Cite this

@article{43895eb2a47841839b122e59014dc92b,

title = "Similarity-based model for ordered categorical data",

abstract = "In a large variety of applications, the data for a variable we wish to explain are ordered and categorical. In this paper, we present a new similarity-based model for the scenario and investigate its properties. We establish that the process is ψ-mixing and strictly stationary and derive the explicit form of the autocorrelation function in some special cases. Consistency and asymptotic normality of the maximum likelihood estimator of the model{\textquoteright}s parameters are proven. A simulation study supports our findings. The results are applied to the Netflix data set, comprised of a survey on users{\textquoteright} grading of movies.",

keywords = "Consistency, ergodicity, mixing, ordered probit, similarity, stationarity",

author = "Gabi Gayer and Offer Lieberman and Omer Yaffe",

note = "Publisher Copyright: {\textcopyright} 2017, {\textcopyright} 2017 Taylor \& Francis Group, LLC.",

year = "2019",

month = mar,

day = "16",

doi = "10.1080/07474938.2017.1308054",

language = "אנגלית",

volume = "38",

pages = "263--278",

journal = "Econometric Reviews",

issn = "0747-4938",

publisher = "Taylor and Francis Ltd.",

number = "3",

}

TY - JOUR

T1 - Similarity-based model for ordered categorical data

AU - Gayer, Gabi

AU - Lieberman, Offer

AU - Yaffe, Omer

PY - 2019/3/16

Y1 - 2019/3/16

N2 - In a large variety of applications, the data for a variable we wish to explain are ordered and categorical. In this paper, we present a new similarity-based model for the scenario and investigate its properties. We establish that the process is ψ-mixing and strictly stationary and derive the explicit form of the autocorrelation function in some special cases. Consistency and asymptotic normality of the maximum likelihood estimator of the model’s parameters are proven. A simulation study supports our findings. The results are applied to the Netflix data set, comprised of a survey on users’ grading of movies.

AB - In a large variety of applications, the data for a variable we wish to explain are ordered and categorical. In this paper, we present a new similarity-based model for the scenario and investigate its properties. We establish that the process is ψ-mixing and strictly stationary and derive the explicit form of the autocorrelation function in some special cases. Consistency and asymptotic normality of the maximum likelihood estimator of the model’s parameters are proven. A simulation study supports our findings. The results are applied to the Netflix data set, comprised of a survey on users’ grading of movies.

KW - Consistency

KW - ergodicity

KW - mixing

KW - ordered probit

KW - similarity

KW - stationarity

UR - https://www.scopus.com/pages/publications/85019244463

U2 - 10.1080/07474938.2017.1308054

DO - 10.1080/07474938.2017.1308054

M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???

AN - SCOPUS:85019244463

SN - 0747-4938

VL - 38

SP - 263

EP - 278

JO - Econometric Reviews

JF - Econometric Reviews

IS - 3

ER -

Similarity-based model for ordered categorical data

Abstract

Bibliographical note

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this