An Experimental Study of Expectation Formation

Abstract

This paper reports on an experimental study of expectation formation-and revision in a time series context. In an adaptive expectations framework, it is shown that the speed of adjustment seems to fall in turning point periods. Expectations are considered as probability density functions, and a scoring system is devised and employed that gives subjects an incentive to report a measure of the dispersion of these functions. This measure, which is inversely related to the confidence with which expectations are held, seems to be inversely related to past forecasting performance. THIS PAPER REPORTS an empirical exploration of the way individuals form and hold expectations about future values of time series variables. In the application of economic models in which expectations about the future play a major role in determining behavior, these expectations are rarely directly observable, and the econometrician is generally forced to assume that a technical rule generates expectations as a simple function only of past observations. One way to see what sort of technical rules make sense in such applications might be to attempt to use this indirect approach to discriminate among possible functional forms. Usually, however, this is computationally burdensome and not terribly revealing. Another approach, currently receiving attention, involves direct analysis of realworld expectations data.2 A third approach, and the ope followed here, is to create and analyze an experimental situation in which the rule followed must be technical because no information other than the past history of the time series in question is available. The main reason for the attractiveness of the experimental approach here, however, lies in the two aspects of expectation formation with which this study is principally concerned. The first of these concerns the influence of turning points in a time series context. The basic hypothesis is due to F. M. Fisher [9, p. 48]: