Experience based dynamic choice: A revealed preference approach

Abstract

We use the revealed preference method to derive a model of dynamic choice where the agent's past experience may influence her current decisions. Our model generalizes the classical individual choice model which is rationalized by utility maximization, and reduces to that model in the absence of experience. As the agent gains experience her utility changes but only in a very restricted fashion. Every period, after an alternative is chosen, the utility of that, and only that alternative, may change while the utility of all other alternatives remains fixed. The model provides a platform on which many behavioral dynamic phenomena may be examined. We utilize it and look into the behavioral implications of bounded memory, status quo bias and variety seeking.

Do Past Experience and Group Heterogeneity Matter to Consumer Preferences? Evidence From a Choice Experiment in Urban China

“…In addition, there are increasing research studies on how familiarity with markets or valuation mechanisms affect the stated or revealed WTP measures (e.g., List, 2001 ; Lejarraga and Gonzalez, 2011 ; Maltz, 2016 ), while other streams of empirical literature are interested in identifying how learning about goods affects preference uncertainty and consumer demand (e.g., Ackerberg, 2003 ; Crawford and Shum, 2005 ; Osborne, 2011 ). However, currently, there is no widely accepted way to test and control the impact and presence of experience and learning on WTP.…”

“…Such real-world decisions occur repeatedly, but the outcomes of those decisions and their likelihoods often have to be inferred from past experience (e.g., Myers and Sadler, 1960 ; Edwards, 1961 ; Katz, 1964 ). Introspections also show that our preferences and choices often change as we gain experience in a particular market, even if other variables related to choice remain constant ( Maltz, 2016 ).…”

What can we learn from experience? An impact analysis of experience on households' preferences for microfinance

“…According to these unobservable inputs, people do not always know exactly how much they value a good and even do not fully incorporate information that tend to contradict their prior choices (Kuhnen and Knutson, 2011;McGee, 2013;Matsumoto and Spence, 2016). As a result, consumers' attitudes, buying intentions and preferences rely heavily on experiences, even when descriptive information is available to them (Lejarraga and Gonzalez, 2011;Claret et al, 2014;Maltz, 2016;Huang and Du, 2018). Second, "preference" is the result of a series of dynamic and experienced choices.…”

Agent-based modeling insights into the optimal distribution of the Fresh Fruit and Vegetable Program

“…The behavioral economics literature finds that past experiences can have an effect on current preferences. Maltz (2016) ) presents a model of preference formation where the utility function is updated based on consumption in the previous period. This framework is broad and can accommodate behavioral assumptions such as brand loyalty, status quo bias, variety seeking, etc.…”