Forecast Accuracy and Economic Gains from Bayesian Model Averaging Using Time Varying Weights

Journal of Forecasting (Impact Factor: 0.93). 08/2010; 29(1-2):251-269. DOI: 10.2139/ssrn.1435232
Source: RePEc


Several Bayesian model combination schemes, including some novel approaches that simultaneously allow for parameter uncertainty, model uncertainty and robust time-varying model weights, are compared in terms of forecast accuracy and economic gains using financial and macroeconomic time series. The results indicate that the proposed time-varying model weight schemes outperform other combination schemes in terms of predictive and economic gains. In an empirical application using returns on the S&P 500 index, time-varying model weights provide improved forecasts with substantial economic gains in an investment strategy including transaction costs. Another empirical example refers to forecasting US economic growth over the business cycle. It suggests that time-varying combination schemes may be very useful in business cycle analysis and forecasting, as these may provide an early indicator for recessions. Copyright © 2009 John Wiley & Sons, Ltd.

Download full-text


Available from: Herman Van Dijk, Jan 13, 2016
  • Source
    • "This model has several features that can be empirically weighted: for example , the economic model suggests that the Great Ratios (e.g., consumption to income, investment to income) are stationary, that only investment-speci…c technology shocks have permanent e¤ects on the real investment good price, and only technology shocks a¤ect productivity in the long run. Model uncertainty derives from uncertainty over Newbold and Harvey 2001, Terui and van Dijk 2002, Hoogerheide, Kleijn, Ravazzolo, van Dijk and Verbeek 2010 and Wright 2008). Some explanation for this phenomenon in particular cases was provided by Hendry and Clements (2002). "
    [Show abstract] [Hide abstract]
    ABSTRACT: The empirical support for features of a Dynamic Stochastic General Equilibrium model with two technology shocks is valuated using Bayesian model averaging over vector autoregressions. The model features include equilibria, restrictions on long-run responses, a structural break of unknown date and a range of lags and deterministic processes. We find support for a number of features implied by the economic model and the evidence suggests a break in the entire model structure around 1984 after which technology shocks appear to account for all stochastic trends. Business cycle volatility seems more due to investment specific technology shocks than neutral technology shocks.
    Full-text · Article · Feb 2013 · International Economic Review
  • Source
    • "Our approach may be related to the work on forecast combination (Bates and Granger, 1969; Deutsch et al., 1994; Guidolin and Timmermann, 2009; Granger and Ramanathan, 1984; Hoogerheide et al., 2010, from a Bayesian viewpoint; Kang, 1986; LeSage and Magura, 1992; Palm and Zellner, 1992; Timmermann, 2006), dimension reduction (Poncela et al., 2011); periodic ARMA models (Basawa and Lund, 2001; Boswijk and Franses, 1996; Francq et al., 2011; Franses and Paap, 2004; Herwartz, 1999; Jones and Brelsford, 1967; Novales and Flores de Frutos, 1997; Osborn and Smith, 1994), panel-data methods (Issler and Lima, 2009). "

    Full-text · Dataset · Dec 2012
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: 1 Contents Foreword Executive summary
    Full-text · Dataset · Jan 2010
Show more