Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

The design of online auctions and markets.

It gives some insight on how we can design bidding and pricing systems to ensure that people only ever attempt to bid what they truly think an item is worth, given what they know and what they think other bidders know.

The broader goals are to establish when these systems might be "incentive compatible" (truthful bidding) and whether or not these results are robust to variations on the assumed information structure (assumptions governing what agents know, what they think other agents know, and so on).

For the HN crowd, it seems like the related field of algorithmic game theory is more practical. It focuses mainly on the implementation and complexity of various auction pricing mechanisms.

The section on 'coordination games and conventions' is also quite interesting and related to problems in distributed computing. For instance, byzantine fault tolerance and common knowledge of global state shares many similarities with economic systems in which many dispersed agents with subjective beliefs about the 'true state' of the world wish to come to some consensus over some market equilibria or economic fundamentals such as interest rates or market prices in the presence of faulty communication and/or strategic trading behaviours.

In these cases, game theory may provide insights on more robust fault tolerant designs, and economic policies that lead to stabler, and more efficient markets.



Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: