Forecasting papers 2009-08-19

In this issue we have Comparing forecasts of Latvia's GDP, Forecast Accuracy and Economic Gains from Bayesian Model Averaging using Time Varying Weights, Forecasting credit growth rate in Romania, No Arbitrage Fractional Cointegration Analysis Of The Range Based Volatility.

  • Comparing forecasts of Latvia's GDP using simple seasonal ARIMA models and direct versus indirect approach
    Date: 2009-08-06
    By: Bušs, Ginters
    This paper contributes to the literature by comparing predictive accuracy of one-period real-time simple seasonal ARIMA forecasts of Latvia's Gross Domestic Product (GDP) as well as by comparing a direct forecast of Latvia's GDP versus three kinds of indirect forecasts. Four main results are as follows. Direct forecast of Latvia's Gross Domestic Product (GDP) seems to yield better precision than an indirect one. AR(1) model tends to give more precise forecasts than the benchmark moving-average models. An extra regular differencing appears to help better forecast Latvia's GDP in an economic downturn. Finally, only AR(1) gives forecasts with better precision compared to a naive Random Walk model.
    Keywords: real-time forecasting; seasonal ARIMA; Direct versus indirect forecasting; Latvia's GDP
    JEL: C13
  • Forecast Accuracy and Economic Gains from Bayesian Model Averaging using Time Varying Weights
    Date: 2009-07-16
    By: Lennart Hoogerheide (Erasmus University Rotterdam)
    Richard Kleijn (PGGM, Zeist)
    Francesco Ravazzolo (Norges Bank)
    Herman K. van Dijk (Erasmus University Rotterdam)
    Marno Verbeek (Erasmus University Rotterdam)
    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.
    Keywords: forecast combination; Bayesian model averaging; time varying model weights; portfolio optimization; business cycle
    JEL: C11
  • Forecasting credit growth rate in Romania: from credit boom to credit crunch?
    Date: 2009-07-26
    By: Albulescu, Claudiu Tiberiu
    The specialists paid a special attention to credit growth in the transitions countries due to its sharp increase during the last years. However, once the financial crisis started in 2008, the credit activity evolution reversed. Consequently, forecasting the credit trend has become a subject of interest in the context of the present financial and economic conditions, because the credit market blockage has a negative impact on economic activity revival and leads to the amplification of the uncertainty on financial markets. The main objective of this paper is to highlight the recent credit developments in Romania and to predict their future evolution. Based on the credit growth rate endogenous factors and using a stochastic simulation econometric model, we demonstrate that this economy experiences a passage from a credit boom to a severe credit crunch. The forecasting exercise results show a credit activity contraction up t! o the end of 2009, demolishing the expectations related to a near economic recovery in Romania.
    Keywords: credit growth rate; forecasts; stochastic simulation; credit crunch
    JEL: C53
  • A No Arbitrage Fractional Cointegration Analysis Of The Range Based Volatility
    Date: 2009-07-15
    By: Eduardo Rossi (Dipartimento di economia politica e metodi quantitativi, University of Pavia, Italy.)
    Paolo Santucci de Magistris (Dipartimento di economia politica e metodi quantitativi, University of Pavia, Italy)
    The no arbitrage relation between futures and spot prices implies an analogous relation between futures and spot volatilities as measured by daily range. Long memory features of the range-based volatility estimators of the two series are analyzed, and their joint dynamics are modeled via a fractional vector error correction model (FVECM), in order to explicitly consider the no arbitrage constraints. We introduce a two-step estimation procedure for the FVECM parameters and we show the properties by a Monte Carlo simulation. The out-of-sample forecasting superiority of FVECM, with respect to competing models, is documented. The results highlight the importance of giving fully account of long-run equilibria in volatilities in order to obtain better forecasts.
    Keywords: Range-based volatility estimator, Long memory, Fractional cointegration, Fractional VECM, Stock Index Futures
    JEL: C32
  • Taken from the NEP-FOR mailing list edited by Rob Hyndman.