Detecting Behavioral Biases in Mixed Human-Proxy Online Auction Markets

Detecting Behavioral Biases in Mixed Human-Proxy Online Auction Markets

Roumen Vragov (Department of Mathematics and Information Technology, Mount Saint Mary College, Newburgh, NY, USA)
DOI: 10.4018/ijsita.2013100104
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

Currently many auction websites directly or indirectly provide support for the use of automated proxies or agents. Buyers can use proxies to monitor auctions and bid at the appropriate time and with the appropriate bid price, sellers can use proxies to set prices or negotiate deals. Proxy complexity varies, however most proxies first require some input on the part of the human trader and then perform the trading task autonomously. This paper proposes and tests a theoretical model of human behavior that can be used to detect behavioral biases in electronic market environments populated by humans and software agents. The paper also quantifies the effect of these biases on individual and business profits.
Article Preview

1. Introduction

During the last decade the number of E-commerce sites has exploded, and buying and selling of goods on the Internet has become a nontrivial task. Just looking at eBay’s listings on 10/04/2013, 11:04 am EST revealed 160,285 auctions for digital cameras going on simultaneously. The cameras come in a wide variety of type, resolution, quality, and they were represented by a variety of vendors. The same digital cameras can be bought at other auction web sites or purchased directly from the sites of the camera makers or from the websites of a few big retailers. Additional units might be posted as available on sites that offer free ads as Craig’s list. Each of these sites sells items in different ways. They require different user information and apply different restrictions. Navigating through this maze of information is still challenging. It is up to the user to choose which web site will better serve her purpose and what her optimal strategy will be given the different rules and requirements of the chosen sites. To deal with this issue programmers have started creating software applications to assist the user in her interactions on-line. These programs have been called robots, auctionbots, software agents, etc. Such a feature is presently available on eBay, under the name of proxy bidding (Roth & Ockenfels, 2002). After a buyer has decided to participate in an auction, she can give a limit price to her proxy. eBay keeps this information private. The proxy then bids in the auction. Every time the proxy is outbid but the current price remains below the limit price provided by the buyer, the proxy bids a minimum increment over the current bid. Meanwhile the buyer does not have to follow the auction. She can devote her time to more important activities. eBay’s director for customer relations in Australia declared that the proxy-bidding service provided by his company’s web-site was flexible enough to allow buyers and sellers to compensate for some inefficiencies due to eBay’s auction design (Davidson, 2005). According to Hayne et al., (2003) 75% of all users on eBay use this feature. The remaining 25% usually participate directly or use more complicated proxies provided by vendors other than eBay. The usage of proxies for e-Commerce transactions is quickly becoming popular. The proxies can be used for searching for items to buy/sell; searching for the best web sites for buying/selling items; and devising different bidding or pricing strategies. Researchers at IBM report that “[trading] robots can make more cash than people when they trade commodities” (Graham-Row, 2001). Other studies (e.g. Miller, 2002) suggest that markets populated entirely of robots cannot attain efficient equilibria. At the same time agent research in E-commerce has flourished at different Universities in the US and abroad (Go to http://cftc.gov/PressRoom/PressReleases/pr6683-13) in which it asked stock market participants for input and recommendations regarding the official introduction of computerized trading systems. Finding out more about how people react and interact with such system is clearly an important and current practical question that needs to be answered.

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 8: 4 Issues (2017): 1 Released, 3 Forthcoming
Volume 7: 4 Issues (2016)
Volume 6: 4 Issues (2015)
Volume 5: 4 Issues (2014)
Volume 4: 4 Issues (2013)
Volume 3: 4 Issues (2012)
Volume 2: 4 Issues (2011)
Volume 1: 4 Issues (2010)
View Complete Journal Contents Listing