Learning about variable demand in the long run

This paper studies the problem of a monopoly who is uncertain about the demand it faces and learns about it over time through its pricing experience. The demand curve facing the monopoly is not constant - it changes over time in a Markovian fashion. We characterize the monopoly's optimal policy and inquire how it differs from an informed monopoly's policy. It turns out that, even when the rate at which the demand varies is negligible, the stationary probability with which the monopoly's policy deviates from its informed counterpart is nonnegligible, as long as the discount factor is below 1.