Bidding Strategies in Agent-Based Continuous Double Auctions

verfasst von: Huiye Ma, Ho-fung Leung

Verlag: Birkhäuser Basel

Buchreihe : Whitestein Series in Software Agent Technologies

Enthalten in: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik" , Springer Professional "Wirtschaft"

Einloggen, um Zugang zu erhalten

Über dieses Buch

Online auctions are a platform to trade goods on the Internet. In this context, negotiation capabilities for software agents in continuous double auctions (CDAs) are a central concern. Agents need to be able to prepare bids for and evaluate offers on behalf of the users they represent with the aim of obtaining the maximum benefit for their users.

For the agents, their bids are decided according to some bidding strategy. However, in CDAs, it is a complex decision problem because of the inherent uncertainty and dynamics of the auction market. In this book, we present a new bidding strategy for agents to adopt in CDAs and propose tools to enhance the performance of existing bidding strategies in CDAs. The superior performance of the new bidding strategy as well as the tools presented in this book are illustrated through extensive experiments.

Inhaltsverzeichnis

Frontmatter

Chapter 1. Introduction

Abstract

With the advent of global computer networks, in particular the Internet and the World Wide Web, electronic commerce (e-commerce) has been taking an increasingly important role in many organizations [50]. It provides a faster, cheaper, more personalized, and more agile way for businesses to interact with their customers and their suppliers. Auctions of various kinds are efficient mechanisms to allocate resources in electronic commerce. In this context, online auctions, institutions where goods are traded on the Internet by the process of bidding and allocating through competition, are among the most widely studied and employed means of interaction [5]. Such online auctions are prevalent because they are an efficient and effective method of allocating goods or services [128], [103], [93].

Chapter 2. Agent-Based CDAs and Bidding Strategies

Abstract

The role of agents in auctions is to represent their users, who may be buyers or sellers or the auctioneer, to achieve particular objectives [58], [54], [81], [83], [50]. Although there are many attributes concluded in the literature, some attributes are essential for CDAs we discuss in this book. First of all, we define exactly what we mean by the term “agent” in CDAs. An agent in CDAs is a software package that can be viewed as a delegate of his user to achieve a good performance which usually means a good profit. To this end, an agent must exhibit the following properties:

Autonomy: The agent is capable of making decisions about what actions to take without constantly referring back to his user
Adaptivity: The agent is capable of adjusting himself to environmental conditions based on trading history, etc

Chapter 3. The Adaptive Attitude Bidding Strategy

Abstract

In this chapter, we develop a novel Adaptive Attitude (AA) bidding strategy that agents can use to participate in CDAs. The AA strategy exploits both the shortterm and long-term attitudes of an agent, and utilizes a threshold-based method with heuristic rules (called the α-ω method) in bid determination. Eagerness is defined for the first time. The effectiveness of the strategy is demonstrated by empirically benchmarking it against the main strategies that have been proposed in the literature and this evaluation shows that the AA strategy is superior in a wide range of market situations.

Chapter 4. Soft Asks and Soft Bids

Abstract

There are several bidding strategies proposed in the literature for agents in CDAs. For most bidding strategies, the asks or bids determined are hard and cannot be compromised. However, for human traders, we notice that the decisions are usually soft and adaptive in different situations. Therefore, we believe that integrating softness and adaptivity into the bidding strategies can enhance the performance of agents. In this chapter, soft asks and soft bids are defined. Experimental results confirm that when agents using different bidding strategies make soft compromise in various situations, their performance is improved significantly in general. Eagerness is extended on the basis of eagerness in Chapter 3. Fuzzy sets and fuzzy logic rules are employed to decide the value of eagerness in order to cope with uncertainty in the dynamic market. To guide agents to adopt soft asks or bids in dynamic and unknown markets, an adaptive mechanism is proposed to adjust the degree of softness of soft asks or bids according to eagerness. Experimental results show that agents adopting the adaptive mechanism generally outperform the corresponding agents without the adaptive mechanism.

Chapter 5. Adaptive Judgement of Price Acceptability

Abstract

There are several bidding strategies in the literature for agents in CDAs to employ, any of which can achieve a good performance. Nevertheless, almost none of these strategies judge whether a price is acceptable before the agents calculate their own asks or bids. [53] by He et al. is the only work in the literature in which the judgement of price acceptability is incorporated into the bidding strategy. With the judgement of price acceptability, agents can improve their profit by accepting asks or bids directly or submitting no asks or bids until they have computed them.

Chapter 6. Adaptive Time Strategies

Abstract

A market mechanism is a set of rules that governs interactions among buyers and sellers and determines how to form a deal [113]. In many real-world situations it is essential to conclude a negotiation among agents within a fixed deadline. For example, there are continuous double auctions over the Internet that require the trading period to terminate within 24 hours. Recently, auction designers have researched the effect of deadlines in auction mechanisms and have applied them to many electronic commerce applications, such as eBay and Amazon [26].

Chapter 7. Discussion of the Results

Abstract

Given the results from the previous chapters, one may want to know how these results can be applied in real-life CDA markets. In order to answer this question, we shall in this chapter highlight the differences between the real-life CDA markets and the agent-oriented CDA markets researched in this book. We then provide an overall discussion on the results observed from these agent-oriented CDA markets, focusing on the scenarios where the results are applicable.

Chapter 8. Conclusions and Future Work

Abstract

With the increasing automation of e-commerce, we believe that ever greater amounts of trading will be conducted in online auctions by software agents. However, to make progress in this area, one of the key problems that needs to be addressed is that of developing effective and efficient bidding strategies or enhancing existing strategies that agents can use to achieve their negotiation objective. To this end, we develop novel strategies and general tools for continuous double auctions.

Backmatter

Titel: Bidding Strategies in Agent-Based Continuous Double Auctions
verfasst von: Huiye Ma
Ho-fung Leung
Verlag: Birkhäuser Basel
Electronic ISBN: 978-3-7643-8730-3
Print ISBN: 978-3-7643-8729-7
DOI: https://doi.org/10.1007/978-3-7643-8730-3