Assessing influence of online reputation on sales using a zero-inflated negative binomial model

Min Xu; Qiang Ye; Xiaolin Wang; Mohan Wang

doi:10.1016/j.procs.2017.11.480

Assessing influence of online reputation on sales using a zero-inflated negative binomial model

Min Xu^*
, Qiang Ye
, Xiaolin Wang
, Mohan Wang

^*Corresponding author for this work

Southwestern University of Finance and Economics
School of Management, Harbin Institute of Technology
Heilongjiang University of Science and Technology
Soochow University

Research output: Contribution to journal › Conference article › peer-review

Abstract

Seller Reputation in online platform plays a vital role in consumer purchasing decisions. Many researchers explore the impact of reputation on sales; however, the majority of them consider sales as continuous variable. We in this paper will treat it as count data and focus on how reputation influence sales volume. Taking into account excess zeros and over-dispersion characteristics of sales, we build a zero-inflated negative binomial (ZINB) model to address the issue. We find that reputation will have two types of influences. The results show that reputation will influence both the amount of sales and the probability that sellers always have no sales.

Original language	English
Pages (from-to)	1108-1113
Number of pages	6
Journal	Procedia Computer Science
Volume	122
DOIs	https://doi.org/10.1016/j.procs.2017.11.480
State	Published - 2017
Externally published	Yes
Event	5th International Conference on Information Technology and Quantitative Management, ITQM 2017 - New Delhi, India Duration: 8 Dec 2017 → 10 Dec 2017

Keywords

count data
reputation
sales volume
zero-inflated negative binomial model

Access to Document

10.1016/j.procs.2017.11.480

Cite this

@article{fbf4544c698f40e08849d5b385f9aa73,

title = "Assessing influence of online reputation on sales using a zero-inflated negative binomial model",

abstract = "Seller Reputation in online platform plays a vital role in consumer purchasing decisions. Many researchers explore the impact of reputation on sales; however, the majority of them consider sales as continuous variable. We in this paper will treat it as count data and focus on how reputation influence sales volume. Taking into account excess zeros and over-dispersion characteristics of sales, we build a zero-inflated negative binomial (ZINB) model to address the issue. We find that reputation will have two types of influences. The results show that reputation will influence both the amount of sales and the probability that sellers always have no sales.",

keywords = "count data, reputation, sales volume, zero-inflated negative binomial model",

author = "Min Xu and Qiang Ye and Xiaolin Wang and Mohan Wang",

note = "Publisher Copyright: {\textcopyright} 2017 The Authors. Published by Elsevier B.V.; 5th International Conference on Information Technology and Quantitative Management, ITQM 2017 ; Conference date: 08-12-2017 Through 10-12-2017",

year = "2017",

doi = "10.1016/j.procs.2017.11.480",

language = "英语",

volume = "122",

pages = "1108--1113",

journal = "Procedia Computer Science",

issn = "1877-0509",

publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - Assessing influence of online reputation on sales using a zero-inflated negative binomial model

AU - Xu, Min

AU - Ye, Qiang

AU - Wang, Xiaolin

AU - Wang, Mohan

PY - 2017

Y1 - 2017

N2 - Seller Reputation in online platform plays a vital role in consumer purchasing decisions. Many researchers explore the impact of reputation on sales; however, the majority of them consider sales as continuous variable. We in this paper will treat it as count data and focus on how reputation influence sales volume. Taking into account excess zeros and over-dispersion characteristics of sales, we build a zero-inflated negative binomial (ZINB) model to address the issue. We find that reputation will have two types of influences. The results show that reputation will influence both the amount of sales and the probability that sellers always have no sales.

AB - Seller Reputation in online platform plays a vital role in consumer purchasing decisions. Many researchers explore the impact of reputation on sales; however, the majority of them consider sales as continuous variable. We in this paper will treat it as count data and focus on how reputation influence sales volume. Taking into account excess zeros and over-dispersion characteristics of sales, we build a zero-inflated negative binomial (ZINB) model to address the issue. We find that reputation will have two types of influences. The results show that reputation will influence both the amount of sales and the probability that sellers always have no sales.

KW - count data

KW - reputation

KW - sales volume

KW - zero-inflated negative binomial model

UR - https://www.scopus.com/pages/publications/85040312363

U2 - 10.1016/j.procs.2017.11.480

DO - 10.1016/j.procs.2017.11.480

M3 - 会议文章

AN - SCOPUS:85040312363

SN - 1877-0509

VL - 122

SP - 1108

EP - 1113

JO - Procedia Computer Science

JF - Procedia Computer Science

T2 - 5th International Conference on Information Technology and Quantitative Management, ITQM 2017

Y2 - 8 December 2017 through 10 December 2017

ER -

Assessing influence of online reputation on sales using a zero-inflated negative binomial model

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this