This paper studies intertemporal pricing policies when selling seasona
l products in retail stores. We first present a continuous time model
where a seller laces a stochastic arrival of customers with different
valuations of the product. For this model, we characterize the optimal
pricing policies as functions of time and inventory. We use this mode
l as a benchmark against which we compare more realistic models that c
onsider periodic pricing reviews. We show that the structure of the op
timal pricing policies in this case is consistent with the procedures
observed in practice; retail stores successively discount the product
during the season and promote a liquidation sale at the end of the pla
nning horizon. We also show that the loss experienced when implementin
g periodic pricing reviews instead of continuous policies is small whe
n the appropriate number of reviews is chosen. Several interesting eco
nomic insights emerge from our analysis. For example, uncertainty in t
he demand for new products leads to higher prices, larger discounts, a
nd more unsold inventory. Finally, we study the effect of announced di
scount policies on prices and profits. We show that stores that have a
dopted this type of strategy usually set prices such that with high pr
obability the merchandise is sold during the first periods and the lar
gest discounts rarely take place.