Multi-agent adaptive systems in dry bulk shipping

Investment decisions in dry bulk shipping form one of the most difficult managerial tasks due to the high degree of uncertainty and the cyclical nature of the market. Adequate information on ship prices is, therefore, crucial when justifying such decisions. This paper is the first to embed trading rules in an evolutionary agent-based system to dynamically incorporate different traders' beliefs on future ship prices. The model is applied to two types of traders, two trading rules and three vessel types for the newbuild and second-hand market in the period 1990-2005. The results indicate that strategy selection is important to understand market pricing. Traders are also shown to benefit from adjusting their strategies over time and over vessel types.