DYNAMIC HIERARCHICAL-MODELS

An analysis of a time series of cross-sectional data is considered und er a Bayesian perspective. Information is modelled in terms of prior d istributions and stratified parametric linear models developed by Lind ley and Smith and dynamic linear models developed by Harrison and Stev ens are merged into a general framework. This is shown to include many models proposed in econometrics and experimental design. Properties o f the model are derived and shrinkage estimators reassessed. Evolution , smoothing and passage of data information through the levels of the hierarchy are discussed. Inference with an unknown scalar observation variance is drawn and an extension to the non-linear case is proposed.