Static and dynamic aspects of the demand for goods in Korea

Abstract

This thesis is concerned with demand analysis for Korea in the context of complete systems of demand equations . Two broad categories of demand analysis - static and dynamic - are considered. This thesis is not only concerned with empirical work but also with the investigation of economic theory relating to the specification of static and dynamic demand systems and also with the development of statistical techniques relating to empirical demand analysis . The scope of this thesis is outlined in the introductory chapter . Chapter 2 sketches the static theory of demand, and reviews various specifications of static demand system. Statistical theories relating to the empirical (static) demand model are considered in Chapter 3, which includes estimation and tests of hypothesis. In respect of estimation, a new algorithm of estimation of a symmetric system(based on analytic solutions of the Lyapunov equation) is developed, and the technique is subsequently extended to the estimation of dynamic demand systems under nonlinear long run equilibrium conditions in Chapter 7. The use of separate induced tests of hypothesis is also considered in Chapter 3. The application of various static demand systems to Korean quarterly household expenditure data is carried out in Chapter 4. Stochastic dynamisation of the disturbances of static demand systems is also considered in chapter 4. Some aspects of dynamic demand theory are examined in Chapter 5. A new specification of a dynamic demand system, based on the dynamic equilibrium assumption, is derived in the context of the Rotterdam model approximation, and applied to the Korean data in the conjunction with dynamic demand analysis. Chapter 6 examines the effect of misspecification of a demand system on tests of demand restrictions, analytically as well as using Monte Carlo simulation. Two specific types of misspecification, (one the includsion of irrelevant explanatory variables, and the other the omission of the relevant explanatory variables) are considered. In chapter 7, problems relating to the estimation of dynamic demand system under long run equilibrium conditions are condidered, and a new algorithm for ML estimation is developed and applied to the examination of the long run demand patterns in Korea. In the last chapter of the thesis, a summary and conclusion are provided.