The goal of this behavioral decision theory project is to develop two tractable models of decision making that weaken standard rationality requirements and accommodate experimental evidence from economics, psychology and marketing.
The first part of the project will develop a new model of random choice; a model in which decision makers have coherent objectives but make errors pursuing those objectives. Describing decision making as random has two important advantages: First, it facilitates the measurement of preferences; that is, random choice models allow the econometrician to quantify the intensity of preference via choice frequencies. Second, random choice models facilitate aggregation across different economic agents. The behavior of distinct individuals can be used as evidence for a single model.
The first part of this research will introduce a new model of random choice, the generalized attribute selection model (GAR), to address key regularities of choice data found in the experiments in economics, psychology and in the marketing literature. We expect this model to become a useful tool for analyzing consumer behavior.
Behavioral models that constrain the decision maker's ability to match her behavior to the details of her environment have proven effective in explaining systematic departures from standard models of "rational choice." The second part of the project will introduce a general analytical device for incorporating behavioral elements into competitive economies. Specifically, this research will analyze a dynamic competitive economy with consumers who are limited in their ability to adjust their consumption choices to their specific circumstances. One application of this theory is asset price volatility: we expect to demonstrate that with a given set of fundamentals (preferences and technology), there is more price volatility in a behavioral competitive equilibrium than in a standard competitive equilibrium.
The ultimate goal of this research is to provide a simple model that permits researchers to quantify and measure the impact of behavioral factors in financial markets and enable them to distinguish the effects of these factors from the effects of fundamentals.
