Non-Stationary Time Series Analysis and Cointegration

This book shows major developments in the econometric analysis of the long run (non-stationary and cointegration) - a field which has developed dramatically over the last twelve years. The papers here describe and evaluate new methods, provide useful overviews, and show detailed implementations helpful to practitioners. Papers include Michael Clements and David Hendry's substantive analysis of economic forecasting, necessarily based around an integral understanding of integration and cointegration. The paper by Fabio Canova, Mary Finn and Adrian Pagan evaluates the real business cycles models using the new techniques. Other topics ocvered include an overview of the different estimators of cointegrating relationships, and a new test of cointegration. Applications are shown finding roots in macroeconomic series, testing the Fisher Hypoethesis, testing money demand functions, to testing for inflationary bubbles. This book provides a good coverage of the depth of this literature showing the importance of an understanding of non-stationarity and cointegration.