Includes bibliographical references and index.

Introduction -- Probability distributions -- Statistical inference -- Two-variable regression analysis -- Estimators and methods of estimation -- The Classical two-variable regression model -- The Classical multiple regression model -- Stochastic explanatory variables -- More about multiple regression -- Non-spherical disturbances -- Estimating dynamic models -- Choosing the appropriate model -- Handling non-stationary time series -- Testing for stationarity -- Cointegration and the estimation of error correction models

Econometrics has experienced remarkable changes in the past 15 years, particularly in the area of time series analysis. The development of cointegration techniques has, for the first time, enabled econometricians to make a serious attempt at dealing with the problems of spurious regressions and non-stationary time series.