Essays on Individual Choice Modeling and Analysis
This dissertation includes three essays. The ﬁrst two essays are concerned with empirical investigation of individual choice by means of laboratory experiment, and the third one is concerned with integrating one particular behavior pattern – consumer learning – into assortment decisions. The summaries of these essays are provided below.
In the ﬁrst essay, we study how customers are choosing prices in a Pay-What-You-Want (PWYW) business model. Under PWYW the price for a product is fully determined by a buyer: the seller cannot reject any oﬀer. We study two factors that can potentially aﬀect PWYW prices: the seller’s production costs and the buyer’s private valuation of the product. We hypothesize that buyers may anticipate the seller’s loss-aversion, so they will be reluctant to choose prices below the costs, and that the prices increase as the buyer’s valuation increases. We suggest a model that incorporates this hypothetical behavior and estimate it using the dataset from a previously published experimental study. Based on the preliminary insights we obtain, we design and conduct our own controlled laboratory experiment. Our ﬁndings suggest that PWYW prices are indeed increasing on average in the buyer’s valuation and seller’s costs.
In the second essay we provide the results of empirical investigation of the behavior of human assortment planners. Assortment planning, that is, the selection of products to oﬀer in a store or the design of a product line for a manufacturer, is often performed by managers without any support from computerized optimization algorithms. The goals of our study are (1) to ﬁnd if human decision makers deviate from the expected proﬁt-maximizing solution in some systematic way and (2) to investigate the eﬀect of decision support tools on the eﬃcacy of assortment planners. To do this, we develop and conduct a behavioral experiment, where the subjects are repeatedly picking assortments in a simplified computerized market environment. We ﬁnd that the subjects perform better when the proﬁt maximizing assortment consists of fewer products and that subjects improve their decisions over time. The eﬀect of decision support is somewhat surprising: under certain conditions providing subjects with more information resulted in worse performance.
In the third essay, we develop a model that explicitly incorporates consumer learning into a ﬁrm’s assortment problem. Consumer’s choice of a product from a particular category is inﬂuenced by her beliefs about how well the product ﬁts her needs. When a consumer purchases a product repeatedly, her experience with it aﬀects her beliefs and, consequently, her future choices. By providing the consumer with an assortment to choose from, the ﬁrm increases the chances that the consumer will ﬁnd a product that is suitable for her and keep purchasing from the ﬁrm in the long term. However, by doing so the ﬁrm faces a risk that a less proﬁtable product gets substituted for a more proﬁtable one. The model we develop allows to investigate this tradeoﬀ analytically.