Forecasting and Information Sharing in Supply Chains Under Quasi-ARMA Demand

In this paper, we revisit the problem of demand propagation in a multi-stage supply chain in which the retailer observes ARMA demand. In contrast to previous work, we show how each player constructs the order based upon its best linear forecast of leadtime demand given its available information. In order to characterize how demand propagates through the supply chain we construct a new process which we call quasi-ARMA or QUARMA. QUARMA is a generalization of the ARMA model. We show that the typical player observes QUARMA demand and places orders that are also QUARMA. Thus, the demand propagation model is QUARMA-in-QUARMA-out. We study the value of information sharing between adjacent players in the supply chain. We demonstrate that under certain conditions information sharing can have unbounded bene¯ts. Our analysis hence reverses and sharpens several previous results in the literature involving information sharing and also opens up many questions for future research.Sy Syms School of Business, Yeshiva University; Department of Information, Operations and Management Science, Stern School of Business, NYU; McCombs School of Business, University of Texas at AustinStatistics Working Papers Serie