DOI QR코드

DOI QR Code

Nonlinearities and Forecasting in the Economic Time Series

  • Published : 2003.12.01

Abstract

It is widely recognized that economic time series involved not only the linearities but also the non-linearities. In this paper, when the economic time series data have the nonlinear characteristics we propose the forecasts method using combinations of both forecasts from linear and nonlinear models. In empirical study, we compare the forecasting performance of 4 exchange rates models(AR, GARCH, AR+GARCH, Bilinear model) and combination of these forecasts for dairly Won/Dollar exchange rates returns. The combination method is selected by the estimated individual forecast errors using Monte Carlo simulations. And this study shows that the combined forecasts using unrestricted least squares method is performed substantially better than any other combined forecasts or individual forecasts.

Keywords

References

  1. 응용통계연구 v.1 no.2 결합예측에 관한 실증적 연구 이우리
  2. 경기대학교 논문집 v.44-1 결합예측 방법에 의한 종합주가지수의 예측 이우리
  3. Oper. Res. Quar. v.20 The Combination of Forecasts Bates,J.M.;Granger,C.W.J. https://doi.org/10.1057/jors.1969.103
  4. JBES, 15-1.43-50 ARCH and Bilinearity as Competing Models for Nonlinear Dependense' Bera,A.K.;Higgins,M.L.
  5. Journal of Econometrics v.31 Generalised Autoregressive Conditional Heteroskedasticity' Bollerslev,T. https://doi.org/10.1016/0304-4076(86)90063-1
  6. JASA v.70 Intervention analysis with applications to economic and environmental problems Box,G.E.P.;Tiao,G.C. https://doi.org/10.1080/01621459.1975.10480264
  7. University of Wisconsin-Madision, Working Paper A Test for Independence Based on the Correlation Dimension Brock,W.A.;Dechert,W.;Scheinkman,J.
  8. Journal of Forecasting v.20 A Double-threshhold GARCH Model for the French Franc/Deutschmarc Exchange Rate Brooks, Chris https://doi.org/10.1002/1099-131X(200103)20:2<135::AID-FOR780>3.0.CO;2-R
  9. JBES v.4 no.1 Combining Economic Forecasts Clemen,R.T.;Winkler,R.L.
  10. Journal of Applied Econometrics v.14 A Monte carlo study of the forecasting performance of empirical SETAR models Clements,M.P.;Smith,J. https://doi.org/10.1002/(SICI)1099-1255(199903/04)14:2<123::AID-JAE493>3.0.CO;2-K
  11. Proceedings of the fourth Berkeley Symposium on Probability and Statistics(Vol.1) v.1 Tests for Separate Families of Hypothesis Cox,D.R.
  12. JRRS, Ser. B v.24 Further Results on Tests of Separate Families of Hypothesis Cox,D.R.
  13. Econometrica v.49 Several Tests for Model Specification in the Presense of Alternative Hypothesis Davidson,R.;Mackinnon,J. https://doi.org/10.2307/1911522
  14. Econometrica v.50 Autoregressive Conditional Heteroskedasticty with Estimates of the Variance of U.K Inflation Engle,R.F. https://doi.org/10.2307/1912618
  15. Econometric Reviews v.5 Modelling the Persistence of Conditional Variances Engle,R.F.;Bollerslev,T. https://doi.org/10.1080/07474938608800095
  16. Econometrica v.55 Estimating Time Varying Risk Premia in the Term Structure: the ARCH-M Model Engle,R.F.;Lilien,D.M.;Robbins,R.P. https://doi.org/10.2307/1913242
  17. J. of International Economics v.47 Linear, nonlinear and essential foreign exchange rate prediction with simple trading rules Gencay,R. https://doi.org/10.1016/S0022-1996(98)00017-8
  18. Journal of Finance v.48 Relationship Betwen the Expected Value and the Volatility of the Nominal Excess Return on Stocks Glosten,L.R.;Jagannathan,R.;Runkle,D. https://doi.org/10.2307/2329067
  19. Journal of Business v.66 no.2 Patterns in three Centuries of Stock Market Prices Goetzmann,W.N. https://doi.org/10.1086/296603
  20. J. of Forecasting v.3 Improved Methods of Combining Forecasts Granger,C.W.J.;Ramanadan,R. https://doi.org/10.1002/for.3980030207
  21. International Economic Review v.33 A Class of Nonlinear ARCH Models Higgins,M.L.;Bera,A.K. https://doi.org/10.2307/2526988
  22. JBES v.7 Modelling heteroskedasticity in daily foreign exchange rates estimation Hsieh,D.A.
  23. Journal of Finance v.46 no.5 Chaos and Nonlinear Dynamics: Application to Financial Markets Hsieh,D. https://doi.org/10.2307/2328575
  24. Biometrika v.72 A Tukey nonadditivity-type test for time series nonlinearity Keenan,D. https://doi.org/10.1093/biomet/72.1.39
  25. J of Time Series Analysis v.4 Diagnostic checking ARMA time series models using squared residual autocorrelations McLeod,A.I.;W.K.Li https://doi.org/10.1111/j.1467-9892.1983.tb00373.x
  26. International Journal of Forecasting v.18 A comparision of the accurcy of short term foreign exchange forecasting methods Meade, Nigel https://doi.org/10.1016/S0169-2070(01)00111-X
  27. JRSS v.31 Tests for specification errors in classical linear least-squares regression analysis Ramsey,J.B.
  28. International Journal of Forecasting v.17 Nonlinearities, cyclical behaviour and predictability in stock markets: international evidence Sarantis, Nicholas https://doi.org/10.1016/S0169-2070(01)00093-0
  29. Management Science v.36 Combining forecasts: operational adjustments to theoretically optimal rules Schmittlein,D.c.;Kim,J.;Morrison,D.G. https://doi.org/10.1287/mnsc.36.9.1044
  30. Quadratic ARCH Models, Review of Economic Studies v.62 Sentana,E. https://doi.org/10.2307/2298081
  31. International Journal of Forecasting v.18 Combined forecasts from linear and nonlinear time series models Terui,Nobuhiko;Herman K. van Dijk https://doi.org/10.1016/S0169-2070(01)00120-0
  32. Biometrika v.73 Nonlinearity tests for time series Tsay,R. https://doi.org/10.1093/biomet/73.2.461
  33. Mathmetics and Computers in Simulation v.59 Non-linear modelling and forecasting of S&P 500 volatility Verhoeven,P.;Pilgram,B.;McAleer,M.;Mees,A. https://doi.org/10.1016/S0378-4754(01)00411-6
  34. JBES v.4 ARCH and Bilinear Time Series Models: comparision and combination Weiss,A.A.
  35. Computational Statistics & Data Analysis v.38 Using genetic algorithms to parameters (d,r) estimation for threshold autoregressive models Wu, Berin;Chih-Li Chang https://doi.org/10.1016/S0167-9473(01)00030-5
  36. Journal of Economic Dynamics and Control v.18 Threshold Heteroskedasti Models Zakoian,J.M. https://doi.org/10.1016/0165-1889(94)90039-6