The book centers on the problem of measuring business cycles in economic time series Business cycle assessment is crucial to monitoring the economy, as well as to the conduct of macroeconomic policy, but decades of research have shown that the modellingof economic cycles is a frustrating issue As a consequence, the estimation of actual business ожбжг cycles relies on relatively simple filters that remove from the series the long-term trend Over the last decade there has been a convergence towards the use of a filter suggested by Hodrick and Prescott (HP) The present paradigm, in actual practice, is to estimate the cycle with the HP filter applied to X11-seasonally adjusted series The procedure has been the subject of academic criticism that has pointed out the risk of obtaining spurious cycles, mostly associated with the empirical and fixed nature of the filter The filter also is known to present some serious drawbacks such as unstable estimation for recent years so that concurrent analysis is very unreliable, poor performance in early detection of turning points, and noise contamination of the cyclical signal which makes interpretation difficult This book treats business cycle estimation by means of the HP filter as a signal extraction problem It shows how the incorporation of straightforward model-based time series tools improves significantly the performance of the filter in terms of the stability of recent-year estimation, of early detection of turning points (and decrease in the number of false alarms), and of smoothness of the cyclical signal The algorithm to enforce the Modified Hodrick-Prescott (MHP) filter is described in detail The authors show that the MHP filter turns out to yield the optimal estimator of the cycle in an unobserved components model where the series consists of a long-term trend, a cycle, a seasonal, and an irregular component The model incorporates ad-hoc features that reflect the desirable properties of the HP filter and series-dependent features that ensure compatibility with the particular structure observed in the series (thus avoiding spurious results) In addition, the existence of an underlying model facilitates diagnostics and inference Regina Kaiser is Professor at the Department of Statistics and Econometricsof the Universidad Carlos III de Madrid, and holds a Ph D in Economics from the European University Institute in Florence, Italy She has published work in the field of applied time series analysis Agustín Maravall is Chief Economist at the Research Department of the Banco de España, and has formerly been Professor at the European University Institute in Florence, Economist at the US Federal Reserve Board of Governors, and Engineer at the Spanish Ministry of Agriculture He holds a Ph D in Economics from the University of Wisconsin-Madison, and a Doctorate in Agricultural Engineering from the University of Madrid, has been Associate Editor of many journals, and has published mostly in the area of applied time series analysis and dynamic econometrics He is a member of the International Statistical Institute, a Fellow of the Journal of Econometrics, and a Fellow of the American Statistical Association.  Содержание:ISBN 0387951121.