| Summary: | This dissertation consists of two essays in behavioral and experimental economics. The first essay provides a theoretical framework that explains the phenomenon of choice overload under irrational preferences. In this theory, the representative consumer not only makes "cognitive errors" but also may have incomplete preferences when making pairwise comparisons across alternatives in a choice set. Expanding a choice set may strengthen cognitive errors and thus increase the likelihood of violating transitivity of preferences. As a result, the representative consumer would find it more difficult to identify a dominant choice. Also, even a constant probability of an incomplete pairwise comparison would lead to a higher probability that the representative consumer is unable to identify a dominant choice when there are more alternatives. Hence, choice overload may arise due to the two potential mechanisms: the expected utility from randomly choosing a non-outside option from the larger choice set is lower than that from the smaller choice set; there exists welfare loss due to the gap between the utility of a dominant choice and the expected utility from randomly choosing a non-outside option. The result of an on-line experiment lends reasonable support to the theory but gives little supportive evidence on the mechanism due to cognitive or search costs, which is heavily endorsed in past literature.
The second essay presents a testing procedure for multiple hypothesis testing in experimental economics. Empiricism in the sciences allows us to test theories, formulate optimal policies, and learn how the world works. In this manner, it is critical that our empirical work provides accurate conclusions about underlying data patterns. False positives represent an especially important problem, as vast public and private resources can be misguided if we base decisions on false discovery. This study explores one especially pernicious influence on false positives---multiple hypothesis testing (MHT). While MHT potentially affects all types of empirical work, we consider three common scenarios where MHT influences inference within experimental economics: jointly identifying treatment effects for a set of outcomes, estimating heterogeneous treatment effects through subgroup analysis, and conducting hypothesis testing for multiple treatment conditions. Building upon the work of Romano and Wolf (2010), we present a correction procedure that incorporates the three scenarios, and illustrate the improvement in power by comparing our results with those obtained by the classic studies due to Bonferroni (1935) and Holm (1979). Importantly, under weak assumptions, our testing procedure asymptotically controls the familywise error rate -- the probability of one false rejection -- and is asymptotically balanced. We showcase our approach by revisiting the data reported in Karlan and List (2007), to deepen our understanding of why people give to charitable causes.
|
|---|
