New forecasting papers 2008-09-10

In this issue we have: Averaging forecasts from VARs with uncertain instabilities ; Improving forecast accuracy by combining recursive and rolling forecasts ; Tests of equal predictive ability with real-time data ; The cyclical component factor model ; Tourism Forecasting: Accuracy of Alternative Econometric Models Revisited ; Modelling and Forecasting Multivariate Realized Volatility.

 

  • Averaging forecasts from VARs with uncertain instabilities
    Date: 2008
    By: Todd E. Clark
    Michael W. McCracken
    URL: http://d.repec.org/n?u=RePEc:fip:fedlwp:2008-030&r=for
    Recent work suggests VAR models of output, inflation, and interest rates may be prone to instabilities. In the face of such instabilities, a variety of estimation or forecasting methods might be used to improve the accuracy of forecasts from a VAR. The uncertainty inherent in any single representation of instability could mean that combining forecasts from a range of approaches will improve forecast accuracy. Focusing on models of U.S. output, prices, and interest rates, this paper examines the effectiveness of combining various models of instability in improving VAR forecasts made with real-time data.
    Keywords: Econometric models ; Economic forecasting
  •  

  • Improving forecast accuracy by combining recursive and rolling forecasts
    Date: 2008
    By: Todd E. Clark
    Michael W. McCracken
    URL: http://d.repec.org/n?u=RePEc:fip:fedlwp:2008-028&r=for
    This paper presents analytical, Monte Carlo, and empirical evidence on combining recursive and rolling forecasts when linear predictive models are subject to structural change. Using a characterization of the bias-variance tradeoff faced when choosing between either the recursive and rolling schemes or a scalar convex combination of the two, we derive optimal observation windows and combining weights designed to minimize mean square forecast error. Monte Carlo experiments and several empirical examples indicate that combination can often provide improvements in forecast accuracy relative to forecasts made using the recursive scheme or the rolling scheme with a fixed window width.
    Keywords: Economic forecasting ; Econometric models
  •  

  • Tests of equal predictive ability with real-time data
    Date: 2008
    By: Todd E. Clark
    Michael W. McCracken
    URL: http://d.repec.org/n?u=RePEc:fip:fedlwp:2008-029&r=for
    This paper examines the asymptotic and finite-sample properties of tests of equal forecast accuracy applied to direct, multi-step predictions from both non-nested and nested linear regression models. In contrast to earlier work in the literature, our asymptotics take account of the real-time, revised nature of the data. Monte Carlo simulations indicate that our asymptotic approximations yield reasonable size and power properties in most circumstances. The paper concludes with an examination of the real-time predictive content of various measures of economic activity for inflation.
    Keywords: Economic forecasting ; Real-time data
  •  

  • The cyclical component factor model
    Date: 2008-09-02
    By: Christian M. Dahl
    Henrik Hansen
    John Smidt (School of Economics and Management, University of Aarhus, Denmark)
    URL: http://d.repec.org/n?u=RePEc:aah:create:2008-44&r=for
    Forecasting using factor models based on large data sets have received ample attention due to the models' ability to increase forecast accuracy with respect to a range of key macroeconomic variables in the US and the UK. However, forecasts based on such factor models do not uniformly outperform the simple autoregressive model when using data from other countries. In this paper we propose to estimate the factors based on the pure cyclical components of the series entering the large data set. Monte Carlo evidence and an empirical illustration using Danish data shows that this procedure can indeed improve on pseudo real time forecast accuracy.
    Keywords: Factor model, Cyclical components, Estimation, Real time forecasting
    JEL: C13 C22 C52
  •  

  • Tourism Forecasting: Accuracy of Alternative Econometric Models Revisited
    Date: 2008-08-13
    By: Haiyan Song
    Egon Smeral (WIFO)
    Gang Li
    Jason L. Chen
    URL: http://d.repec.org/n?u=RePEc:wfo:wpaper:y:2008:i:326&r=for
    This study evaluates the forecasting accuracy of five alternative econometric models in the context of predicting the quarterly international tourism demand in 25 countries or country groupings. Tourism demand is measured in terms of tourist expenditure by inbound international visitors in a destination. Two univariate time series models are included in the forecasting comparison as benchmarks. Accuracy is assessed in terms of error magnitude. Seasonality is an important feature of forecasting models and requires careful handling. For each of the 25 destinations, individual models are estimated over the 1980Q1-2005Q1 period, and forecasting performance is assessed using data covering the 2005Q2-2007Q1 period. The empirical results show that the time-varying parameter (TVP) model provides the most accurate short-term forecasts, whereas the naïve (no-change) model performs best in long-term forecasting up to two ye! ars. This study provides new evidence of the TVP model's outstanding performance in short-term forecasting. Through the incorporation of a seasonal component into the model, the TVP model forecasts short-run seasonal tourism demand well.
    Keywords: tourism forecasting, econometric models, time series models, forecasting accuracy
  •  

  • Modelling and Forecasting Multivariate Realized Volatility
    Date: 2008-09-02
    By: Roxana Chiriac
    Valeri Voev (School of Economics and Management, University of Aarhus, Denmark)
    URL: http://d.repec.org/n?u=RePEc:aah:create:2008-39&r=for
    This paper proposes a methodology for modelling time series of realized covariance matrices in order to forecast multivariate risks. The approach allows for flexible dynamic dependence patterns and guarantees positive definiteness of the resulting forecasts without imposing parameter restrictions. We provide an empirical application of the model, in which we show by means of stochastic dominance tests that the returns from an optimal portfolio based on the model's forecasts second-order dominate returns of portfolios optimized on the basis of traditional MGARCH models. This result implies that any risk-averse investor, regardless of the type of utility function, would be better-off using our model.
    Keywords: Forecasting, Fractional integration, Stochastic dominance, Portfolio optimization, Realized covariance
    JEL: C32 C53 G11
  • Taken from the NEP-FOR mailing list edited vby Rob Hyndman.