Price Prediction in a Trading Agent Competition

TL;DR

This paper analyzes various hotel price prediction methods used in the 2002 Trading Agent Competition, highlighting the effectiveness of game-specific information and machine learning techniques in improving prediction accuracy and agent performance.

Contribution

It provides a comprehensive survey of prediction approaches in TAC-02, evaluates their relative efficacy, and introduces a new measure linking prediction accuracy to game performance.

Findings

Game-specific information improves prediction accuracy.

Machine learning effectively models price relationships.

Analytical methods can match machine learning accuracy without historical data.

Abstract

The 2002 Trading Agent Competition (TAC) presented a challenging market game in the domain of travel shopping. One of the pivotal issues in this domain is uncertainty about hotel prices, which have a significant influence on the relative cost of alternative trip schedules. Thus, virtually all participants employ some method for predicting hotel prices. We survey approaches employed in the tournament, finding that agents apply an interesting diversity of techniques, taking into account differing sources of evidence bearing on prices. Based on data provided by entrants on their agents' actual predictions in the TAC-02 finals and semifinals, we analyze the relative efficacy of these approaches. The results show that taking into account game-specific information about flight prices is a major distinguishing factor. Machine learning methods effectively induce the relationship between flight and hotel prices from game data, and a purely analytical approach based on competitive equilibrium analysis achieves equal accuracy with no historical data. Employing a new measure of prediction quality, we relate absolute accuracy to bottom-line performance in the game.