Learning, Experimentation, and the Optimal Output Decisions of a Competitive Firm

This paper considers the effect of learning from experience on the output decisions of a perfectly competitive firm faced with the demand uncertainty. Specifically, a Bayesian framework for expectations formation and demand forecasting by a perfectly competitive firm is presented. Focusing the analysis on the determination of optimal sequential output decisions, it is shown that through output experimentation, the experimenting firm will select a non-myopic sequential policy and will tend to overproduce. The exact magnitude of the overproduction and the economic value of experimentation are contingent upon model parameters and the length of the planning horizon.

• 저널Management Science · 28(6) · 589–603
• 토픽Economic theories and models · Economics and Econometrics
• DOI10.1287/mnsc.28.6.589
• 저자Giora Harpaz, Wayne Lee, Robert L. Winkler