NEW INSIGHTS INTO MODELING CHOICE BEHAVIOR UNDER UNCERTAINTY

Abstract

This dissertation explores alternative ways to understand choice preferences and explain data better. Understanding choice behavior has important policy implications. Choices are dependent on numerous factors, some of which are uncertain and unobservable to the researcher. To model choice behavior under uncertainty, researchers make several assumptions regarding individuals’ cognitive thought processes, the functional form of utility, and which behavioral anomalies to incorporate. In this research, the choice behavior of three distinct sets of stakeholders is investigated. The first chapter examines location choices made by recreational fishermen using an alternative model called the Case-Based Decision Theory (CBDT). This model captures the thinking process of a decision-maker based on the similarity of circumstances. CBDT hypothesizes that decision-makers rely on stored memory, past experience, and analogical reasoning to make choices. Fishermen tend to be biased towards qualitatively assessing their alternative locations based on their intuition and experience rather than numerically estimating the expected value. As a result, we find that CBDT outperforms the conventional Linear Additive Model when comparing both in and out of sample fits. The second chapter investigates variety-seeking and habit forming behavior exhibited by birdwatchers when it comes to choosing bird watching sites. Birdwatchers unlike fishermen are variety-seekers in site choice preferences. Variety-seeking behavior makes it difficult to predict choice preferences and therefore difficult to identify any change in site preference when there is a policy change. This chapter introduces a two-stage model based on the framework adapted from CBDT to capture this effect. With this model, we find a statistically significant combined effect for variety-seeking and habit forming among birdwatchers. This approach to predict choice behavior by agents using case-based reasoning in my first two chapters has been observed in several empirical settings, however, it has never been applied in a natural resource or an environmental context or used in non-market valuation studies. The final chapter investigates how introducing a visual representation of policy alternatives regarding a local dam affects the choice preferences of residents in the city where the dam is located. This study uses a split sample labeled choice experiment to describe five possible future alternatives for the dam via text, images with text, and video with text. Previous studies support the theory that visualizing the available choices help better comprehend information. Drawing from this conjecture, we find that certain dam alternatives have a relatively higher preference when images are introduced while alternatives such as dam removal