Robust Firm Pricing with Panel Data

We present an econometric framework for robust pricing based on credible assumptions about consumer behavior. First, we introduce a panel data discrete choice model whose realistic assumptions about consumer behavior result in partial identification of preferences. Next we address the resulting ambiguity in the firm pricing problem since it cannot use standard techniques to maximize expected profit. We present four models diciphering in how they account for potential variation in preferences over time.We then use the minimax regret criterion as a decision-making rule for firms facing ambiguity about consumer preferences. We study monopoly and oligopoly. We perform simulations that illustrate our methodology and compare our model results to the most common parametric model used in this setting. We show that in the likely event that the parametric assumptions do not represent the underlying data generating process, decisions based on the models presented in this paper substantially improve firm performance.