This authoritative collection brings together the most important papers in time series econometrics published since 1990. These articles cover a range of central aspects of the field, concentrating in the main on theoretical and methodological developments. Taken together, they provide an overview of the current status of research in time series econometrics, emphasising those areas that appear to have attracted most recent interest in the profession. Volume I includes sections on unit root and stationarity tests; cointegration; structural breaks; nonlinearity; and long memory. Volume II covers conditional heteroskedasticity; stochastic volatility; unobserved components; trend function analysis; prediction; seasonality; and causality. These volumes will be essential reading for all who have an interest in this rapidly advancing subject.