Forecasting papers 2009-01-20

In this issue we have: Forecasting Exchange Rates with a Large Bayesian VAR ; Yield-Curve Based Probit Models for Forecasting U.S. Recessions ; Out-of-sample comparison of copula specifications in multivariate density forecasts ; and more.

  • Forecasting Exchange Rates with a Large Bayesian VAR
    Date: 2008
    By: A. Carriero
    G. Kapetanios
    M. Marcellino
    URL: http://d.repec.org/n?u=RePEc:eui:euiwps:eco2008/33&r=for
    Models based on economic theory have serious problems at forecasting exchange rates better than simple univariate driftless random walk models, especially at short horizons. Multivariate time series models suffer from the same problem. In this paper, we propose to forecast exchange rates with a large Bayesian VAR (BVAR), using a panel of 33 exchange rates vis-a-vis the US Dollar. Since exchange rates tend to co-move, the use of a large set of them can contain useful information for forecasting. In addition, we adopt a driftless random walk prior, so that cross-dynamics matter for forecasting only if there is strong evidence of them in the data. We produce forecasts for all the 33 exchange rates in the panel, and show that our model produces systematically better forecasts than a random walk for most of the countries, and at any forecast horizon, including at 1-step ahead.
    Keywords: Exchange Rates, Forecasting, Bayesian VAR
    JEL: C53 C11 F31
  • Yield-Curve Based Probit Models for Forecasting U.S. Recessions: Stability and Dynamics
    Date: 2008-05
    By: Heikki Kauppi (Department of Economics, University of Turku)
    URL: http://d.repec.org/n?u=RePEc:tkk:dpaper:dp31&r=for
    Recent research provides controversial evidence on the stability of yield-curve based binary probit models for forecasting U.S. recessions. This paper reviews so far applied specifications and presents new procedures for examining the stability of selected probit models. It finds that a yield-curve based probit model that treats the binary response (a recession dummy) as a nonhomogeneous Markov chain produces superior in-sample and out-of-sample probability forecasts for U.S. recessions and that this model specification is stable over time. Thus, the failure of yieldcurve based forecasts to signal the 1990-1991 and 2001 recessions should not be attributed to parameter instability, instead the evidence suggests that these events were inherently uncertain.
    Keywords: recession forecast, yield curve, dynamic probit models, parameter stability
    JEL: C22 C25 E32 E37
  • Out-of-sample comparison of copula specifications in multivariate density forecasts
    Date: 2008-10
    By: Cees Diks (University of Amsterdam)
    Valentyn Panchenko (School of Economics, University of New South Wales)
    Dick van Dijk (Econometric Institute, Erasmus University Rotterdam)
    URL: http://d.repec.org/n?u=RePEc:swe:wpaper:2008-23&r=for
    We introduce a statistical test for comparing the predictive accuracy of competing copula specifications in multivariate density forecasts, based on the Kullback-Leibler Information Criterion (KLIC). The test is valid under general conditions: in particular it allows for parameter estimation uncertainty and for the copulas to be nested or nonnested. Monte Carlo simulations demonstrate that the proposed test has satisfactory size and power properties in finite samples. Applying the test to daily exchange rate returns of several major currencies against the US dollar we find that the Student's t copula is favored over Gaussian, Gumbel and Clayton copulas. This suggests that these exchange rate returns are characterized by symmetric tail dependence.
    Keywords: Copula-based density forecast; semiparametric statistics; out-of-sample forecast evaluation; Kullback-Leibler Information Criterion; empirical copula
    JEL: C12 C14 C32 C52 C53
  • Testing directional forecast value in the presence of serial correlation
    Date: 2008-12
    By: Oliver Blaskowitz
    Helmut Herwartz
    URL: http://d.repec.org/n?u=RePEc:hum:wpaper:sfb649dp2008-073&r=for
    Common approaches to test for the economic value of directional forecasts are based on the classical Chi-square test for independence, Fisher's exact test or the Pesaran and Timmerman (1992) test for market timing. These tests are asymptotically valid for serially independent observations. Yet, in the presence of serial correlation they are markedly oversized as confirmed in a simulation study. We summarize serial correlation robust test procedures and propose a bootstrap approach. By means of a Monte Carlo study we illustrate the relative merits of the latter. Two empirical applications demonstrate the relevance to account for serial correlation in economic time series when testing for the value of directional forecasts.
    Keywords: Directional forecasts, directional accuracy, forecast evaluation, testing independence, contingency tables, bootstrap
    JEL: C32 C52 C53 E17 E27 E47 F17 F37 F47 G11
  • Modeling Electricity Prices: From the State of the Art to a Draft of a New Proposal
    Date: 2008-02
    By: Matteo Manera (University of Milano-Bicocca)
    Massimiliano Serati (Institute of Economics, Cattaneo University – LIUC, Castellanza)
    Michele Plotegher (ENI SPA)
    URL: http://d.repec.org/n?u=RePEc:fem:femwpa:2008.9&r=for
    In the last decades a liberalization of the electric market has started; prices are now determined on the basis of contracts on regular markets and their behaviour is mainly driven by usual supply and demand forces. A large body of literature has been developed in order to analyze and forecast their evolution: it includes works with different aims and methodologies depending on the temporal horizon being studied. In this survey we depict the actual state of the art focusing only on the recent papers oriented to the determination of trends in electricity spot prices and to the forecast of these prices in the short run. Structural methods of analysis, which result appropriate for the determination of forward and future values are left behind. Studies have been divided into three broad classes: Autoregressive models, Regime switching models, Volatility models. Six fundamental points arise: the peculiarities of electr! icity market, the complex statistical properties of prices, the lack of economic foundations of statistical models used for price analysis, the primacy of uniequational approaches, the crucial role played by demand and supply in prices determination, the lack of clearcut evidence in favour of a specific framework of analysis. To take into account the previous stylized issues, we propose the adoption of a methodological framework not yet used to model and forecast electricity prices: a time varying parameters Dynamic Factor Model (DFM). Such an eclectic approach, introduced in the late ‘70s for macroeconomic analysis, enables the identification of the unobservable dynamics of demand and supply driving electricity prices, the coexistence of short term and long term determinants, the creation of forecasts on future trends. Moreover, we have the possibility of simulating the impact that mismatches between demand and supply have over the price variable. This way it is possible! to evaluate whether congestions in the network (eventually leading bl ack out phenomena) trigger price reactions that can be considered as warning mechanisms.
    Keywords: Electricity Spot Prices, Autoregressive Models, GARCH Models, Regime Switching Models, Dynamic Factor Models
    JEL: C2 C3 Q4
  • Industrial Coal Demand in China: A Provincial Analysis
    Date: 2008-02
    By: Matteo Manera (University of Milano-Bicocca)
    Cristina Cattaneo (Fondazione Eni Enrico Mattei, Milan and University of Sussex)
    Elisa Scarpa (Edison Trading)
    URL: http://d.repec.org/n?u=RePEc:fem:femwpa:2008.8&r=for
    In recent years, concerns regarding the environmental implications of the rising coal demand have induced considerable efforts to generate long-term forecasts of China's energy requirements. Nevertheless, none of the previous empirical studies on energy demand for China has tackled the issue of modelling coal demand in China at provincial level. The aim of this paper is to fill this gap. In particular, we model and forecast the Chinese demand for coal using time series data disaggregated by provinces. Moreover, not only does our analysis account for heterogeneity among provinces, but also, given the nature of the data, it captures the presence of spatial autocorrelation among provinces using a spatial econometric model. A fixed effects spatial lag model and a fixed effects spatial error model are estimated to describe and forecast industrial coal demand. Our empirical results show that the fixed effect spatial l! ag model better captures the existing interdependence between provinces. This model forecasts an average annual increase in coal demand to 2010 of 4 percent.
    Keywords: Energy demand, Coal demand, China, Spatial econometrics, Panel data, Forecasting
    JEL: C23 E6 Q31 Q41
  • Prediction Accuracy of Different Market Structures – Bookmakers versus a Betting Exchange
    Date: 2008
    By: Egon Franck (Institute for Strategy and Business Economics, University of Zurich)
    Erwin Verbeek (Institute for Strategy and Business Economics, University of Zurich)
    Stephan Nüesch (Institute for Strategy and Business Economics, University of Zurich)
    URL: http://d.repec.org/n?u=RePEc:iso:wpaper:0096&r=for
    There is a well-established literature on separately testing the prediction power of different betting market settings. This paper provides an inter-market comparison of the forecasting accuracy between bookmakers and a major betting exchange. Employing a dataset covering all football matches played in the major leagues of the "Big Five" (England, France, Germany, Italy, Spain) during three seasons (5478 games in total), we find evidence that the betting exchange provides more accurate predictions of the same underlying event than bookmakers. A simple betting strategy of selecting bets for which bookmakers offer lower probabilities(higher odds) than the bet exchange generates above average and, in some cases, even positive returns.
    Keywords: Nprediction accuracy, betting, bookmaker, betting exchange, probit regression
  • Taken from the NEP-FOR mailing list edited by Rob Hyndman.