Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/221317 
Title: 
Year of Publication: 
1991
Series/Report no.: 
Discussion Paper No. 958
Publisher: 
Northwestern University, Kellogg School of Management, Center for Mathematical Studies in Economics and Management Science, Evanston, IL
Abstract: 
I report the results of nine experimental asset market sessions. The traded assets were contingent claims on two "states" with known state probabilities and identical aggregate payoffs across states. Since subjects could diversify away all idiosyncratic risks, this results in prices predicted to equal expected values regardless of risk preferences. IN addition, no-arbitrage restrictions lead to precise predictions for the sum of the individual asset prices. Thus, in these single-period asset markets, price bubbles are well defined without knowing risk preferences. Bubbles regularly form. Subjects seldom exploit the resulting arbitrage opportunity. Bubbles arise even when subjects have gained experience in 15 periods, when subjects can trade directly in the "unit portfolio" of all contingent claims and when subjects can sell claims short. However, direct portfolio trading may force subjects to recognize the assets' interdependence and price them accordingly. Expected utility maximizing models cannot explain these results. Non-expected models (for example, models of decision regret) may explain them.
Document Type: 
Working Paper

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.