A Memory-Based Model of Bounded Rationality

Abstract

In order to investigate the impact of limited memory on human behavior, I develop a model of memory grounded in psychological and biological research. I assume that people take their memories as accurate and use them to make inferences. The resulting model predicts both over- and underreaction but provides enough structure to predict when each effect dominates. I then use this framework to study the consumption decision. The results match empirical work on consumption predictability as well as differences in the marginal propensity to consume from different income streams. Most importantly, because it ties the extent of bias to a measurable aspect of the stochastic process being forecasted, the model makes testable empirical predictions.