Learning Buyer Behavior under Realistic Pricing Restrictions

We propose a new sample efficient algorithm to learn the parameters governing the purchasing behavior of a utility maximizing buyer, who responds to prices, in a repeated interaction setting. The key feature of our algorithm is that it can work with arbitrary price constraints that the seller may impose. This overcomes a major shortcoming of previous approaches, which use unrealistic prices to learn these parameters making them unsuitable in practice. Once the relevant parameters are learned by our algorithm, in addition to being useful for decisions such as pricing goods to maximize the seller’s profit, the models can also be used for other decision problems such as inventory control and promotion planning.