Case-based Decision Theory (CBDT) suggests that decisions under ucnertainty are made by analogies to previously-encountered problems. The theory postulates a similarity function over decision problems and a utility functionon outcomes, such that acts are evaluated by a similarity-weighted sum of the utility othey yielded in past cases in which they were chosen. It gives rise to the concept of "aspiration level" in the following sense: if this level is attained by some acts, the decision will only choose among them, and will not even experiment withothers. Thus a case-based decision maker may be "satisficed" with a choice, and will not maximize his/her utility function even if the "same" problem is encountered over and over again. In this paper we discuss the process by which the aspiration level is updated. An adjustment rule is "realistic" if the aspiration level is (almost always) set to be an average of its previous value and the best average-performance so far encountered. It is "ambitious" if at least one of the following holds: (i) the initial aspiration level is set at a high level, or (ii) the aspiration level is set to exceed the maximal average performance by some constant infinitely often. While we propose realistic-but-ambitious adjustment rules for decision under uncertainty at large, we focus here on the case in which the decision maker is repeatedly faced with the "same" problem, assuming that each choice yields an independent realization of a given random variable. We show that if the adjustment rule is realistic-but-ambitious in the sense of (i), then with arbitrarily high probability the decision maker will asymptotically choose only expected-utility maximizing acts. Ambitiousess in the sense of (ii) above guarantees the same result with probability 1, and for all underlying payoff distributions. Hence, case-based decision makers who are both ambitious and realistic will "learn"
(This abstract was borrowed from another version of this item.)
References listed on IDEAS
Please report citation or reference errors to , or , if you are the registered author of the cited work, log in to your RePEc Author Service profile, click on "citations" and make appropriate adjustments.:
- M. Kandori & G. Mailath & R. Rob, 1999.
"Learning, Mutation and Long Run Equilibria in Games,"
Levine's Working Paper Archive
500, David K. Levine.
- Kandori, Michihiro & Mailath, George J & Rob, Rafael, 1993. "Learning, Mutation, and Long Run Equilibria in Games," Econometrica, Econometric Society, vol. 61(1), pages 29-56, January.
- Kandori, M. & Mailath, G.J., 1991. "Learning, Mutation, And Long Run Equilibria In Games," Papers 71, Princeton, Woodrow Wilson School - John M. Olin Program.
- Gilboa, Itzhak & Schmeidler, David, 1995.
"Case-Based Decision Theory,"
The Quarterly Journal of Economics,
MIT Press, vol. 110(3), pages 605-39, August.
- Itzhak Gilboa & David Schmeidler, 1993. "Case-Based Consumer Theory," Discussion Papers 1025, Northwestern University, Center for Mathematical Studies in Economics and Management Science.
When requesting a correction, please mention this item's handle: RePEc:eee:gamebe:v:15:y:1996:i:1:p:1-26. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Zhang, Lei)
If references are entirely missing, you can add them using this form.