Theory and applications of disequilibrium econometrics

Abstract

In this thesis further consideration is given to the theory and application of disequilibrium models in econometrics. Since the important work by Fair and Jaffee (1972) there has been a substantial body of literature dealing with this topic. Typically much of this research has been devoted to increasing knowledge about the estimation techniques for these models and applications are less common. An analysis of the Official Short Term Money Market attempts to redress this imbalance and to show how a disequilibrium approach has the potential to enrich the specification of a model. However, before proceeding with this application a number of theoretical issues arc considered. Firstly, the economic framework of disequilibrium models is considered, especially where it has an influence on the econometric framework. It is a weakness in some applications that closer attention is not paid to the economic aspects of disequilibrium in the markets under consideration. While estimation of models of markets in disequilibrium is well established in the literature, there arc some issues not dealt with entirely adequately, and where this is the case an attempt at resolution is made. Some of these issues lead on to a simulation study of three stage instrumental variables estimators. Also examination of estimation techniques indicate problems in estimating some disequilibrium specifications and so the Score test is examined as a means of avoiding estimation of disequilibrium models whenever possible, by using a test based on equilibrium estimates. Finally, attention is focussed on specification and estimation of a model of the Official Short Term Money Market. Compared to other applications of disequilibrium in individual markets this model is somewhat larger, with disaggregation of supply and introduction of unplanned behaviour being features peculiar to the model. The problem of hypothesis testing in the model is also given special attention. The results obtained give some support to the use of a disequilibrium approach in a monthly model of the Official Short Term Money Market.