A Kalman Filter Approach to Analyze Multivariate Hedonic Pricing in Dynamic Supply-Chain Markets

Accurate forecasting of market price developments is essential in achieving excellent market performance. Especially in oligopolistic markets for durable consumer products a robust understanding of selling prices is important, as it drives pricing behavior as well as procurement, inventory and production decisions. Moreover, a supply chain perspective is indispensable for pricing forecasts since companies not only compete for product sales but also for limited resources. Our paper explores the use for price forecasting of multivariate hedonics-based pricing models that explicitly model selling prices with the market valuation of constituting parts. The model is applied to the TAC, a trading agent competition in computer market. To find unknown component prices series we had used the Kalman filter technique to smooth and forecast implicit prices using the EM algorithm. This approach is based on a simple recursive procedure for estimating the parameters of our state space model by maximum of expectation of likelihood. Finally, we present results of our analysis to establish the viability of this method and a simulation on TAC data to generate implicit prices, a possible indicator for taking decisions.