MarketLink currently supports asset markets with a single asset. The asset can be long-lived, as in the "asset market bubble" experiment, or an asset may expire at the conclusion of a period, as in the information aggregation experiment.

Asset market bubble experiment

The asset market experiment that we have included in MarketLink is based on the asset market "bubble" experiment by Smith, Suchanek, and Williams [1988]. In this market, traders begin a multi-period experiment session with units of an asset that pay a dividend in each period which is random, but drawn from a known distribution. Traders also have a currency endowment which they can use to purchase units of the asset. The experiment frequently produces a result that is surprising and contrary to standard asset pricing models. A risk neutral trader should be willing to pay exactly the expected per period dividend of the asset times the number of periods remaining in the experiment session. Yet the asset price frequently begins below this risk neutral asset value. As the sesssion proceeds, the asset price eventually exceeds the risk neutral value. Finally, as the end of the session approaches, the asset value crashes to its fundamental value.

To add the Asset Market Bubble Experiment configuration to your experiment profile, click on the button below.

Information Aggregation Experiment

The configuration below is used to run an experiment with three states. The two trader types in the experiment have dividends that differ across the states. As a result of the differences in their dividend schedules, the two trader types have different expected dividends, so we expect some exchange between them.

This experiment is based on an experiment in Plott and Sunder (1988). In their experiment the three states were called X, Y, and Z . When state Z occurred, some of the traders were told that Y had not occurred, and some of the traders were told that X had not occurred. Consequently, taken together, these traders could determine that state Z occurred. If information were somehow aggretated in this way, the price of the dividend should be equal to the dividend received in state Z by the type that has the highest dividend in that state.

The experiment that we have included here doesn't allow private information, but in the near future we will add that capability.

References

Plott, Charles and Shyam Sunder (1988). "Rational Expectations and the Aggregation of Diverse Information in Laboratory Security Markets," Econometrica, 56:5, pp. 1085-1118.

Smith, Vernon L., Gerry L. Suchanek, and Arlington W. Williams (1988). "Bubbles, Crashes, and Endogenous Expectations in Experimental Spot Asset Markets," Econometrica, 56:5, pp. 1119-1151.