BAYESIAN-ANALYSIS OF CONCURRENT TIME-SERIES WITH APPLICATION TO REGIONAL IBM REVENUE DATA

Business data frequently arise in the form of concurrent time series. We present a general framework for simultaneous modeling and fitting o f such series using the class of Box-Jenkins models. This framework is an exchangeable hierarchical Bayesian model incorporating dependence among the series. Our motivating data set consists of regional IBM rev enue available monthly for several geographic regions. Stationary seas onal autoregressive models are simultaneously fit to the regional data series using various error covariance specifications for the strong i nter-regional dependence. A modified Gibbs sampling algorithm is used to carry out the fitting and to enable all subsequent inference. Graph ical techniques using predictive distributions are employed to assess model adequacy and to select among models. Outlier estimation and pred iction under the chosen model are used for planning and to measure the effect of special promotional events.