Co-integration, error correction, and the econometric analysis of non-stationary data /
[Book]
Anindya Banerjee [and others].
New York :
Oxford University Press,
1993.
xiii, 329 pages :
illustrations ;
24 cm.
Advanced texts in econometrics
Includes bibliographical references (pages 311-321) and indexes.
Introduction and Overview --- Equilibrium relationships and the long run --- Stationarity and equilibrium relationships --- Equilibrium and the specification of dynamic models --- Estimation of long-run relationships and testing for orders of integration and co-integration --- Preliminary concepts and definitions --- Data representation and transformations --- Examples: typical ARMA processes --- Empirical time series: money, prices, output, and interest rates --- Outline of later chapters --- Appendix --- Linear Transformations, Error Correction, and the Long Run in Dynamic Regression --- Transformations of a simple model --- The error-correction model --- Bardsen and Bewley transformations --- Equivalence of estimates from different transformations --- Homogeneity and the ECM as a linear transformation of the ADL --- Variances of estimates of long-run multipliers --- Expectational variables and the interpretation of long-run solutions --- Properties of Integrated Processes --- Spurious regression --- Trends and random walks --- Some statistical features of integrated processes --- Asymptotic theory for integrated processes --- Using Wiener distribution theory --- Near-integrated processes --- Testing for a Unit Root --- Similar tests and exogenous regressors in the DGP --- General dynamic models for the process of interest --- Non-parametric tests for a unit root --- Tests on more than one parameter --- Further extensions --- Asymptotic distributions of test statistics --- Co-integration --- Polynomial matrices --- Integration and co-integration: formal definitions and theorems --- Significance of alternative representations --- Alternative representations of co-integrated variables: two examples --- Engle-Granger two-step procedure --- Regression with Integrated Variables --- Unbalanced regressions and orthogonality tests --- Dynamic regressions --- Functional forms and transformations --- Appendix: Vector Brownian Motion --- Co-integration in Individual Equations --- Estimating a single co-integrating vector --- Tests for co-integration in a single equation --- Response surfaces for critical rabies --- Finite-sample biases in OLS estimates --- Powers of single-equation co-integration tests --- An empirical illustration --- Fully modified estimation --- A fully modified least-squares estimator --- Dynamic specification --- Appendix: Covariance Matrices --- Co-integration in Systems of Equations --- Co-integration and error correction --- Estimating co-integrating vectors in systems --- Inference about the co-integration space --- An empirical illustration --- A second example of the Johansen maximum likelihood approach --- Asymptotic distributions of estimators of co-integrating vectors in I(1) systems --- Conclusion.
0
This book is wide-ranging in its account of literature on cointegration and the modelling of integrated processes (those which accumulate the effects of past shocks). Data series which display integrated behaviour are common in economics, although techniques appropriate to analyzing such data are relatively new, with few existing expositions of the literature. This book explores relationships among integrated data series and their use in dynamic econometric modelling. The concepts of cointegration and error-correction models are fundamental components of the modelling strategy. This area of time series econometrics has grown in importance over the past decade and is of interest to both econometric theorists and applied econometricians. By explaining the important concepts informally and presenting them formally, the book bridges the gap between purely descriptive and purely theoretical accounts of the literature. The work describes the asymptotic theory of integrated processes and uses the tools provided by this theory to develop the distributions of estimators and test statistics. It emphasizes practical modelling advice and the use of techniques for systems estimation. A knowledge of econometrics, statistics, and matrix algebra at the level of a final-year undergraduate or first-year undergraduate course in econometrics is sufficient for most of the book. Other mathematical tools are described as they occur. -- Publisher description.