Research Regional Economy Money and Payments Studies National Economy Global Economy Capital Markets Financial Intermediation Publications Economic Policy Review Current Issues in Economics and Finance Staff Reports/Research Papers   Research Update Publications and Other Research Economists Program for Resident Scholars Visiting Scholars

Home > Research > Research Publications Research Update New Titles in the Staff Reports Series Number 2, 2008 Return to index page   Staff Reports home page   Quantitative Methods   No. 326, May 2008 Dynamic Factor Models with Time-Varying Parameters: Measuring Changes in International Business Cycles Marco Del Negro and Christopher Otrok Del Negro and Otrok develop a dynamic factor model with time-varying factor loadings and stochastic volatility in both the latent factors and idiosyncratic components. They employ this new measurement tool to study the evolution of international business cycles in the post–Bretton Woods period, using a panel of output growth rates for nineteen countries. The authors find: 1) statistical evidence of a decline in volatility for most countries, with the timing, magnitude, and source (international or domestic) of the decline differing across countries; 2) some evidence of a decline in business cycle synchronization for Group of Seven countries, but otherwise no evidence of changes in synchronization for the sample countries, including European and euro-area countries; and 3) convergence in the volatility of business cycles across countries.   No. 327, May 2008 Revisiting Useful Approaches to Data-Rich Macroeconomic Forecasting Jan J. J. Groen and George Kapetanios Groen and Kapetanios revisit a number of data-rich prediction methods that are widely used in macroeconomic forecasting and compare them with a lesser known alternative: partial least squares regression. The authors provide a theorem showing that when the data comply with a factor structure, principal components and partial least squares regressions provide asymptotically similar results. They also argue that forecast combinations can be interpreted as a restricted form of partial least squares regression. The study applies partial least squares, principal components, and Bayesian ridge regressions to a large panel of monthly U.S. macroeconomic and financial data to forecast CPI inflation, core CPI inflation, industrial production, unemployment, and the federal funds rate across different subperiods and finds that partial least squares regression usually has the best out-of-sample performance when compared with the two other data-rich ­prediction methods.