Computational Economics

, Volume 26, Issue 1, pp 65–89 | Cite as

Detecting Business Cycle Asymmetries Using Artificial Neural Networks and Time Series Models

  • Khurshid M. Kiani


This study examines possible existence of business cycle asymmetries in Canada, France, Japan, UK, and USA real GDP growth rates using neural networks nonlinearity tests and tests based on a number of nonlinear time series models. These tests are constructed using in-sample forecasts from artificial neural networks (ANN) as well as time series models.

Our study results based on neural network tests show that there is statistically significant evidence of business cycle asymmetries in these industrialized countries. Similarly, our study results based on a number of time series models also show that business cycle asymmetries do prevail in these countries. So we are not able to evaluate the impact of monetary policy or any other shocks on GDP in these countries based on linear models.


B22 C32 C45 E32 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Andrews, D. (2001). Testing when parameter is under the boundary of the maintained hypothesis. Econometrica, 69, 683–734.CrossRefGoogle Scholar
  2. Axelrod, R. (1987). The evolution of the strategies in the iterated prisoner’s dilemma. In L.D. Davis (ed.), Algorithm and Simulated Annealing, Los Altos, CA: Morgan Kaufmann, pp. 32–41.Google Scholar
  3. Balke, N. and Fomby, T. (1994). Large shocks, small shocks and economic fluctuations: Outliers in macroeconomic time series. Journal of Applied Econometrics, 9, 181–200.Google Scholar
  4. Beaudry, P. and Koop, G. (1993). Do recessions permanently change output? Journal of Monetary Economics, 31, 149–163.CrossRefGoogle Scholar
  5. Bidarkota, P. (1999). Sectoral investigation of asymmetries in the conditional mean dynamics of the real US GDP. Studies in Nonlinear Dynamics and Econometrics, 3, 191–200.Google Scholar
  6. Bidarkota, P. (2000). Asymmetries in the conditional mean dynamics of real GNP: Robust evidence. The Review of Economics and Statistics, 82, 153–157.CrossRefGoogle Scholar
  7. Box, G. and Jenkins, G. (1976). Time Series Analysis, Forecasting and Control, Holden-Day, San Francisco, CA.Google Scholar
  8. Brunner, A. (1992). Conditional asymmetries in real GNP: A semi-nonparametric approach. Journal of Business & Economics Statistics, 10, 65–72.Google Scholar
  9. Brunner, A. (1997). On the dynamic properties of asymmetric models of real GNP. The Review of Economics and Statistics, 79, 321–326.CrossRefGoogle Scholar
  10. Davis, R. (1977). Hypothesis testing when a nauisance parameter is present only under the alternative. Biometrika, 64, 247–254.Google Scholar
  11. De Janog, K. (1975). An Analysis of the Behavior of a Class of Genative Adaptive System, Unpublished PhD Dissertation. University of Michigan.Google Scholar
  12. DeLong, J. and Summers, L. (1986). Are business cycles symmetrical? In R. Gordon (ed.), The American Business Cycle: Continuity and Change, University of Chicago Press, Chicago.Google Scholar
  13. Diebold, F. and Rudebusch, G. (1990). A nonparametric investigation of duration dependence in the American business cycle. Journal of Political Economy, 98, 596–616.CrossRefGoogle Scholar
  14. Dunteman, G. (1989). Principal Components Analysis, Sage, Newbury Park.Google Scholar
  15. Dorsey, R. and Walter, J. (1995). Algorithm for estimation problems with multiple optima, nondifferentiability, and other irregular features. Journal of Business Economics and Statistics, 13, 53–66.Google Scholar
  16. Eitrheim, O. and Terasvirta, T. (1996). Testing adequacy of smooth transition autoregressive models. Journal of Econometrics, 74, 59–75.CrossRefMathSciNetGoogle Scholar
  17. Falk, B. (1986). Further evidence on the asymmetric behavior of economic time series over the business cycle. Journal of Political Economy, 94, 1096–1109.CrossRefGoogle Scholar
  18. French, M. and Sichel, D. (1993). Cyclical patterns in the variance of economic activity. Journal of Business and Economic Statistics, 11, 113–119.Google Scholar
  19. Granger, C. (1995). Modeling nonlinear relationships between extended-memory variables. Econometrica, 63, 265–279.Google Scholar
  20. Goldberg, D. (1989). Genetic Algorithm in Search, Optimization and Machine Learning’s, Addison Wisely, Reading, MA.Google Scholar
  21. Hornik, K., Stinchcombe, M. and White, H. (1990). Universal approximation of an unknown mapping and its derivatives using multilayer feed forward neural networks. Neural Networks, 3, 551–560.CrossRefGoogle Scholar
  22. Hornik, K., Stinchcombe, M. and White, H. (1989). Multilayer feed forward networks are universal approximators. Neural Networks, 2, 359–366.CrossRefGoogle Scholar
  23. Kang, S. (1991). An Introduction of the Use of Feed Forward Neural Networks for Forecasting. Ph.D. Thesis, Kent State University.Google Scholar
  24. Judge, G., Griffiths, Hill, C., Lutkepohl, H. and Lee, T. (1985). The Theory and Practice of Econometrics, 2nd edn., Wiley: New York.Google Scholar
  25. Kastens, T., Featherstone, and Biere, W. (1995). A neural network premier for agricultural economists. Agricultural Finance Review, 55, 54–73.Google Scholar
  26. Keenan, D. (1985). A Tucky nonlinear type test for time series nonlinearities. Biometrika, 72, 39–44.Google Scholar
  27. Kiani, K. and Bidarkota, P. (2004). On business cycle asymmetries in G7 countries. Oxford Bulletin of Economics and Statistics, 66, 333–353.CrossRefGoogle Scholar
  28. Kydland, F. and Prescott, E. (1982). Time to build and aggregate fluctuations. Econometrica, 50, 1345–1370.Google Scholar
  29. Lee, T., White, H. and Granger, C. (1993). Testing for neglected nonlinearity. Journal of Econometrics, 56, 269–290.CrossRefGoogle Scholar
  30. Long, J. and Plosser, C. (1983). Real business cycles. Journal of Political Economy, 91, 39–69.CrossRefGoogle Scholar
  31. Lippmann, R. (1987). An introduction to computing with neural nets. IEEE ASSP Magazine, April, 4–22.Google Scholar
  32. Lucas, R. (1976). Econometric policy evaluation: A critique. Carnegie-Rochester Conference Series in Public Policy, 1, 19–46.Google Scholar
  33. Luukkonen, R., Saikkonen, P. and Terasvirta, T. (1988). Testing linearity against smooth transition autoregressive models. Biometrika, 75, 491–499.Google Scholar
  34. Marimon, R., McGratten, E. and Sergeant, T. (1990). Money as medium of exchange in an economy with artificially intelligent agents. Journal of Economics Dynamics and Control, 14, 329–373.CrossRefGoogle Scholar
  35. Neftci, S. (1984). Are economic time series asymmetric over the business cycle? Journal of Political Economy, 92, 307–328.CrossRefGoogle Scholar
  36. Politis, D. and Romeo, J. (1994). Large sample confidence region based on sub-sample under minimal assumption. Annals of Statistics, 22, 2031–2052.Google Scholar
  37. Politis, D., Romeo, J. and Wolf, M. (1997). Sub-sampling for heteroskedastic time series. Journal of Econometrics, 81, 281–317.CrossRefGoogle Scholar
  38. Potter, S. (1995). A non-linear approach to GNP, U.S. Journal of Applied Econometrics, 10, 109– 125.Google Scholar
  39. Quenouille, M. (1949). A note on bias in estimation. Biometrika, 43, 353–360.Google Scholar
  40. Ramsey, J. (1969). Tests for specification errors in classical linear least-squares regression analysis. Journal of the Royal Statistical Society, 31, 350–371.Google Scholar
  41. Ramsey, J. and Rothman, P. (1996). Time irreversibility and business cycle asymmetry. Journal of Money Credit and Banking, 28, 1–21.Google Scholar
  42. Reilly, D. and Cooper, L. (1990). An overview of neural networks: Early models to real world systems. In S.F. Zornetzer, J.L. Davis and C. Lau (eds.), An Introduction to Neural and Electronic Networks, Academic Press, New York.Google Scholar
  43. Scheinkman, J. and LeBaron, B. (1989). Non-linear dynamics and GNP data. In W. Barnett et al. (eds.), Economic Complexity: Chaos, Sunspots, Bubbles, and Non-linearity, Cambridge University Press, Cambridge.Google Scholar
  44. Schwerz, G. (1978). Estimating the dimension of a model. Annals of Statistics, 6, 461–464.Google Scholar
  45. Schumpeter, J. (1939). Business Cycles: A Theoretical, Historical, and Statistical Analysis of the Capitalist Process, McGraw-Hill, New York.Google Scholar
  46. Sichel, D. (1989). Are business cycles asymmetric? A correction. Journal of Political Economy, 97, 1255–1260.CrossRefGoogle Scholar
  47. Sirinivasan, D., Liew, A. and Chang, C. (1994). A neural network short-term load forecaster. Electric Power System Research, 28, 227–234.CrossRefGoogle Scholar
  48. Sims, K. (1980). Macroeconomic reality. Econometrica, 48, 1–49.Google Scholar
  49. Stinchcombe, M. and White, H. (1989). Universal approximation using feed forward network with non-sigmoid hidden layer activation functions. In Proceedings of the International Joint Conference on Neural Networks, Washington, DC. Vol. I, IEEE Press, New York, pp. 613–618.Google Scholar
  50. Stock, J. and Watson, M. (1999). Business cycle fluctuations in US macroeconomic time series. In J.B. Taylor and M. Woodford (eds.), Handbook of Macroeconomics, Vol. 1, Elsevier Science, B.V., Amsterdam.Google Scholar
  51. Tang, Z. and Fishwick, A. (1993). Feed forward neural networks as models for time series forecasting. ORSA Journal of Computing, 5, 374–385.Google Scholar
  52. Terasvirta, T. (1994). Specification, estimation and evaluation of smooth transition autoregressive models. Journal of American Statistical Association, 89, 208–218.Google Scholar
  53. Terasvirta, T. and Anderson, H. (1992). Characterizing nonlinearities in business cycles using smooth transition autoregressive models. Journal of Applied Econometrics, 7, S119–S136.Google Scholar
  54. Terasvirta, T., Lin, C. and Granger, C. (1993). Power of neural network test. Journal of Time Series Analysis, 14, 209–220.Google Scholar
  55. Terasvirta, T., Tjostein, D. and Granger, C. (1994). Aspects of modeling nonlinear time series. Handbook of Time Series Analysis, 4, 2975–2937.Google Scholar
  56. Tsay, R. (1986). Non-linearity tests for time series. Biometrika, 73, 461–466.Google Scholar
  57. Tsay, R. (1988). Nonlinear time series analysis of blowfly population. Journal of Time Series Analysis, 9, 247–263.Google Scholar
  58. Tuckey, J. (1958). A bias and confidence in not-quite large samples. Annals of Mathematical Statistics, 29, 614–623. (abstracts)Google Scholar
  59. Sharda, R. (1994). Neural networks for the MS/OR analyst: An application bibliography. Interfaces, 24, 116–130.Google Scholar
  60. Vishwakarma, K. (1994). Recognizing business cycle turning points by means of a neural network. Computational Economics, 7, 175–185.CrossRefGoogle Scholar
  61. White, H. (1989a). Some asymptotic results for learning in single hidden layer feed forward network models. Journal of Statistical Association, 84, 1003–1013.Google Scholar
  62. White, H. (1989b). Learning in artificial neural networks: A statistical perspective. Neural Computations, 1, 425–464.Google Scholar
  63. White, H. (1990). Connectionist nonparametric regression: Multilayer feed forward network can learn arbitrary mappings. Neural Networks, 3, 535–550.CrossRefGoogle Scholar
  64. Wong, F. (1991). Time series forecasting using back propagation using neural networks. Neurocomputing, 2, 147–149.CrossRefGoogle Scholar
  65. Wu, C. (1990). On the asymptotic properties of the jackknife histogram. Annals of Statistics, 18, 1438–1452.Google Scholar
  66. Zhang, G., Patuwo, B. and Hu, M. (1998). Forecasting with artificial neural networks: The state of arts. International Journal of Forecasting, 14, 35–62.CrossRefGoogle Scholar
  67. Ziari, H., Leatham, D. and Ellinger, P. (1997). Developing of statistical discriminant mathematical programming model via re-sampling estimation techniques. American Journal of Economics, 79, 1352–1362.Google Scholar

Copyright information

© Springer Science + Business Media, Inc. 2005

Authors and Affiliations

  • Khurshid M. Kiani
    • 1
  1. 1.Kansas State UniversityU.S.A.

Personalised recommendations