We model a decision maker who observes available alternatives according to a list and stochastically forgets some alternatives. Each time the decision maker observes an item in the list, she recalls previous alternatives with some probability, conditional on those alternatives being recalled until this point. The decision maker maximizes a preference relation over the set of alternatives she can recall. We show that if every available alternative is chosen with strictly positive probability, the preference order and the list order must coincide in any limited memory representation. Under the full support assumption, the preference ordering, the list ordering and the memory parameters are uniquely identified up to the ranking of the two least preferred alternatives. We provide conditions on observable choice probabilities that characterize the model under the full support assumption. We then apply our model to study the pricing problem of a monopolist who faces consumers with limited memory. We show that when the probability of forgetting is high, the monopolist is better off charging a lower price than the optimal price in the perfect memory case.

We study how the allocation of attention to different options and the accessibility of options from memory affect decision making. To distinguish between attention and memory, we propose a two-stage stochastic consideration set formation process. An alternative enters the decision maker’s consideration set if it is investigated in the initial attention stage and it is remembered in the subsequent recall stage. In the initial attention stage, the decision maker investigates each available alternative with some alternative-specific probability. In the recall stage, the decision maker recalls each alternative that she investigated in the attention stage with some probability. The probability of recalling an alternative depends on the memorability of the alternative and its position in the order of investigation in the attention stage. Investigating an alternative more recently enhances the probability of recalling it. The decision maker chooses the option that maximizes her preference relation over her consideration set. Under the assumption that the investigation of alternatives is observable, we provide testable implications and show that the revealed preference, attention parameters and memory parameters can be uniquely identified from observable repeated choices.

Price data is subject to measurement error due to temporary promotions, coupon usage by consumers or bargaining between consumers and sellers. Even when detailed data on transaction prices are available, prices of not purchased alternatives may not be recorded by the dataset, making counterfactual analysis difficult. In this paper, we study consumer behavior when the transaction prices are subject to such variations that are not captured by the data. We show that the weak axiom of revealed preference may lead to erroneous conclusions about violations of utility maximization and incorrectly measure consumer preferences if unobserved price variations are ignored.