J. Andreoni and J. Miller. Auctions with artificial adaptive agents. Games and Economic Behavior, 10:39--64, 1995.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....niches more easily, particularly in the face of competing producers. We model the learning of each producer using a genetic algorithm (GA) We chose to use GAs primarily because of their effectiveness at learning in high dimensional, nonstationary problems. GAs are also commonly used [1, 4] to study the adaptive behavior of agents in a multi agent system. We cannot claim that a GA is the optimal algorithm for this problem, or even that the parameters or encoding we have chosen are ideal. Our goal is to identify the sorts of knowledge that will help a producer locate profitable ....

J. Andreoni and J. Miller. Auctions with artificial adaptive agents. Games and Economic Behavior, 10:39--64, 1995.

....period. The CDA is the dominant institution for real world trading of equities, derivatives, etc. Experiments with human subjects in simulated CDAs find reliable price convergence close to theoretical equilibria [9, 10] Several studies examined CDAs with various computerized bidding agents [1, 2, 4, 5, 8], including a major comparative study in the Santa Fe Double Auction Tournament (SFDAT) 7] The principal conclusion was that Todd Kaplan s simple sniping strategy, which waits for the bid ask Permission to make digital or hard copies of all or part of this work for personal or classroom use ....

J. Andreoni and J. Miller. Auctions with artificial adaptive agents. Games and Economic Behavior, 10:39--64, 1995.

.... 1995) We model such a process which uses only very weak assumptions about the degree of information and rationaltiy on the agents side by a simple genetic algorithm (GA) GAs have been applied to describe the learning behavior of boundeldy rational agents in various economic models (e.g. Andreoni and Miller, 1995; Arifovic, 1994, 1995, 1996a; Dawid, 1996b; Dawid and Mehlmann 1995, Routledge, 1995) It has been shown in several of these papers that this process is well suited to describe boundedly rational behavior and that the behavior generated by this kind of algorithm often fits empirical data gathered ....

Andreoni, J. and Miller, J. H., 1995, 'Auctions with Artificial Adaptive Agents', Games and Economic Behavior, 10, 39-64.

....extreme division of the surplus is also predicted by gametheoretic SPE, whereas the deadline timing is not (SPE outcomes are indifferent on timing here) Notice that this timing in the evolving agents actually corresponds very well to real life situations. 14 A similar approach was applied in [1] for a genetic algorithm. 16 When a risk of breakdown exists, on the other hand, most agreements are reached in the very first round. If the the finite game becomes long enough, the deadline is therefore no longer perceived by the evolving agents. In that case, the evolutionary system deviates ....

J. Andreoni and J.H. Miller. Auctions with artificial adaptive agents. Games and Economic Behavior, 10:39--64, 1995.

.... by imitation (selection reproduction) learning by communication (crossover) and learning by experimentation (mutation) Recent economic research shows that genetic algorithm learning performs quite well as a learning mechanism when applied to some standard benchmark cases of economic theory (Andreoni and Miller (1995), Arifovic (1994, 1995, 1996) Bullard and Duffy (1998) Dawid (1996a) Genetic algorithm learning is able to reproduce the results of at least some mainstream economic models, especially concerning their stability properties. One of the scientific challenges to GA learning research is to find out ....

Andreoni, James and Miller, John H. (1995). Auctions with Artificial Adaptive Agents. Games and Economic Behavior, 10, 39--64.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC