by separating the interface from the core auction procedures. This design provides a responsive interface and tolerates system and network disruptions, but necessitates careful timekeeping procedures to ensure temporal accuracy. The AuctionBot has been used extensively in classroom exercises, and is available

and extending applications. It provides generic programming facilities that are useful for a variety of applications, such as variables and loops and procedures. Fur-thermore, Tcl is embeddable: its interpreter is implemented as a library of C procedures that can easily be incorporated into applications

the theoretical core of the thesis, generalising the expectation-maximisation (EM) algorithm for learning maximum likelihood parameters to the VB EM al-gorithm which integrates over model parameters. The algorithm is then specialised to the large family of conjugate-exponential (CE) graphical models, and several

and the auction issuer and the computational efficiency is very reasonable. This architecture can be used to implement any mechanism design where the important factor is the complexity of the decision procedure.

Employing a procedure suggested by a simple theoretical model of auctions in which bidders and sellers have observable and heterogenous reputations for default, we examine the effect of reputation on price in a data set drawn from the online auction site eBay. Our main empirical result

A benchmark "package auction" is introduced in which bidders may determine their own packages on which to bid. If all bidders bid straightforwardly, then the outcome is a point in the core of the exchange economy that minimizes the seller's revenue. When goods are substitutes

This paper presents a decision theoretic framework that an autonomous agent can use to bid e#ectively across multiple, simultaneous auctions. Specifically, our framework enables an agent to make rational decisions about purchasing multiple goods from a series of auctions that operate di

The goal of environmental policy is to protect the well-being of humans and the ecosystems vital to human existence. Because benefit-cost analyses are now required at the federal level, and increasingly at the state level, much attention has been paid to the development of practical, credible approaches for estimating the benefits and costs of environmental programs. Although policy makers have a good handle on measuring the explicit costs associated with increased environmental protection, at present several disparate approaches are utilized to measure economic values of environmental goods and services. Arguably the most contentious of these approaches is the Contingent Valuation Method (CVM), which allows the researcher to measure the total value of the commodity in question (see Peter A. Diamond and Jerry A. Hausman's [1994] critical review). Chief amongst these concems is whether hypothetical bias is inherent in CVM responses. Some recently published studies provide evidence that suggests important differences exist between responses from real and hypothetical valuation questions.1 Recognizing this shortcoming, researchers have adopted both ex ante and ex post techniques to overcome hypothetical bias. Although these attempts have not been completely successful in dealing with hypothetical

An algorithm is described for solving large-scale instances of the Symmetric Traveling Salesman Problem (STSP) to optimality. The core of the algorithm is a "polyhedral" cutting-plane procedure that exploits a subset of the system of linear inequalities defining the convex hull

, and others. A discussion of the speed performance of the procedure and a proof of the theorem that underhes ~ts completeness are also g~ven. The procedure has been used extensively asthe deductive core of a system for program specification and verification.