Online Learning and Pricing for Multiple Products With Reference Price Effects

We consider the dynamic pricing problem of a monopolist seller who sells a set of mutually substitutable products over a finite time horizon. Customer demand is sensitive to the price of each individual product and the reference price which is formed from a comparison among the prices of all products. To maximize the total expected profit, the seller needs to determine the selling price of each product and also select a reference product (to be displayed) that affects the consumer's reference price. However, the seller initially knows neither the demand function nor the optimal reference product, but can learn them from past observations on the fly. As such, the seller faces the classical trade-off between exploration (learning the demand function and reference price) and exploitation (using what has been learned thus far to maximize revenue). We propose a rate-optimal dynamic learning-and-pricing algorithm that integrates iterative least squares estimation and bandit control techniques in a seamless fashion. We show that the cumulative regret, that is, the expected revenue loss caused by not using the optimal policy over T$$ T $$ periods, is upper bounded by & Otilde;(n2T)$$ \overset{\widetilde }{O}\left({n}<^>2\sqrt{T}\right) $$ where & Otilde;(<middle dot>)$$ \overset{\widetilde }{O}\left(\cdotp \right) $$ hides any logarithmic factors. We also establish the regret lower bound (for any learning policies) to be Omega(n2T)$$ \Omega \left({n}<^>2\sqrt{T}\right) $$. We then generalize our analysis to a more general demand model. Our algorithm performs consistently well numerically, outperforming an exploration-exploitation benchmark. The use of price experimentation and estimation techniques could be readily applied in real retail management.