Hypothesis Testing with Restricted Spectral Density Matrices, with an Application to Uncovered Interest Parity.

This paper explores one technique for estimation and hypothesis testing in dynamic linear models. The method combines the analytics of moving average solutions to dynamic models, together with the computational advantages of the Whittle likelihood. As an application, the hypothesis of uncovered interest parity under rational expectations is represented in the form of cross-equation restrictions and tested using the technique. Copyright 1989 by Economics Department of the University of Pennsylvania and the Osaka University Institute of Social and Economic Research Association.