Empirical Economics

, Volume 43, Issue 1, pp 399–426 | Cite as

Understanding forecast failure of ESTAR models of real exchange rates

  • Daniel BuncicEmail author


The forecast performance of the empirical ESTAR model of Taylor et al. (2001) is examined for 4 bilateral real exchange rate series over an out-of-sample evaluation period of nearly 12 years. Point as well as density forecasts are constructed, considering forecast horizons of 1 to 22 steps head. The study finds that no forecast gains over a simple AR(1) specification exist at any of the forecast horizons that are considered, regardless of whether point or density forecasts are utilised in the evaluation. Non-parametric methods are used in conjunction with simulation techniques to learn about the models and their forecasts. It is shown graphically that the nonlinearity in the conditional means (or point forecasts) of the ESTAR model decreases as the forecast horizon increases. The non-parametric methods show also that the multiple steps ahead forecast densities are normal looking with no signs of bi-modality, skewness or kurtosis.


Purchasing power parity Regime modelling Non-linear real exchange rate models ESTAR Forecast evaluation Density forecasts Non-parametric methods 

JEL Classification

C22 C52 C53 F31 F47 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Amisano G, Giacomini R (2007) Comparing density forecasts via weighted likelihood ratio tests. J Bus Econ Stat 25(2): 177–190CrossRefGoogle Scholar
  2. Bao Y, Lee TH, Saltoğlu B (2007) Comparing density forecast models. J Forecast 26(3): 203–225CrossRefGoogle Scholar
  3. Breunig RV, Najarian S, Pagan AR (2003) Specification testing of Markov switching models. Oxford Bullet Econ Stat 65(S1): 703–725CrossRefGoogle Scholar
  4. De Gooijer JG, Kumar K (1992) Some recent developments in non-linear time series modelling, testing, and forecasting* 1. Int J Forecast 8(2): 135–156CrossRefGoogle Scholar
  5. Diebold FX, Mariano RS (1995) Comparing predictive accuracy. J Bus Econ Stat 13(1): 253–263Google Scholar
  6. Eitrheim Ø, Teräsvirta T (1996) Testing the adequacy of smooth transition autoregressive models. J Econ 74(1): 59–75Google Scholar
  7. Franses PH, van Dijk D (2000) Nonlinear time series models in empirical finance. Cambridge University Press, CambridgeCrossRefGoogle Scholar
  8. Gallant AR (1987) Nonlinear statistical models. John Wiley, New YorkCrossRefGoogle Scholar
  9. Giacomini R, White H (2006) Tests of conditional predictive ability. Econometrica 74(6): 1545–1578CrossRefGoogle Scholar
  10. Granger CWJ, Teräsvirta T (1993) Modelling nonlinear economic relationships. Oxford University Press, OxfordGoogle Scholar
  11. Harvey DI, Leybourne SJ, Newbold P (1997) Testing the equality of prediction mean squared errors. Int J Forecast 13(2):281–291CrossRefGoogle Scholar
  12. Lundbergh S, Teräsvirta T (2002) Forecasting with smooth transition autoregressive models. In: Clements MP, Hendry DF (eds) A Companion to economic forecasting. Blackwell Publishers, Oxford pp 485– 509Google Scholar
  13. McCracken MW, West KD (2002) Inference about predictive ability. In: Clements MP, Hendry DF (eds) A companion to economic forecasting. Blackwell Publishers, Oxford, pp 299–321Google Scholar
  14. Mitchell J, Hall SG (2005) Evaluating, comparing and combining density forecasts using the KLIC with an application to the Bank of England and NIESR fan charts of inflation. Oxford Bull Econ Stat 67(S1): 995–1033CrossRefGoogle Scholar
  15. Newey WK, West KD (1987) A simple, positive semi-definite, heteroskedasticity and autocorrelation consistent covariance matrix. Econometrica 55(3): 703–708CrossRefGoogle Scholar
  16. Obstfeld M, Taylor AM (1997) Nonlinear aspects of goods-market arbitrage and adjustment: Hecksher’s commodity points revisited. J Japanese Int Econ 11(4): 441–479CrossRefGoogle Scholar
  17. Pagan AR (2002) Learning about models and their fit to data. Int Econ J 16(2):1–18Google Scholar
  18. Pagan AR, Ullah A (1999) Nonparametric Econometrics. Cambridge University Press, New YorkGoogle Scholar
  19. Ramsey JB (1996) If nonlinear models cannot forecast, what use are they. Studies Nonlin Dynamic Econ 1(2): 65–86Google Scholar
  20. Rapach DE, Wohar ME (2006) The out-of-sample forecasting performance of nonlinear models of real exchange rate behavior. Int J Forecast 22(2):341–361CrossRefGoogle Scholar
  21. Scott DW (1992) Multivariate density estimation: theory, practice, and visualization. John Wiley and Sons, New York, ChichesterCrossRefGoogle Scholar
  22. Silverman BW (1986) Density estimation for statistics and data analysis. Chapman and Hall, LondonGoogle Scholar
  23. Taylor MP, Peel DA, Sarno L (2001) Nonlinear mean-reversion in real exchange rates: Towards a solution to the purchasing power parity puzzles. Int Econ Rev 42(4): 1015–1042CrossRefGoogle Scholar
  24. Teräsvirta T (1994) Specification, estimation, and evaluation of smooth transition autoregressive models. J Am Stat Assoc 89(425):208–218Google Scholar
  25. Teräsvirta T (2006) Forecasting economic variables with nonlinear models. In: Elliott G, Granger C, Timmermann A (eds) Handbook of economic forecasting, vol 1. Elsevier, Amsterdam, pp 413–457Google Scholar
  26. Teräsvirta T, van Dijk D, Medeiros MC (2005) Smooth transition autoregressions, neural networks, and linear models in forecasting macroeconomic time series: A re-examination. Int J Forecast 21(4): 755–774CrossRefGoogle Scholar
  27. van Dijk D, Franses PH (2003) Selecting a nonlinear time series model using weighted tests of equal forecast accuracy. Oxford Bull Econ Stat 18(S1): 727–744CrossRefGoogle Scholar
  28. Zhang G, Patuwo BE, Hu MY (1998) Forecasting with artificial neural networks: the state of the art. Int J Forecast 14(1):35–62CrossRefGoogle Scholar

Copyright information

© Springer-Verlag 2011

Authors and Affiliations

  1. 1.University of St. Gallen, Institute of Mathematics and StatisticsSt. GallenSwitzerland

Personalised recommendations