Dating Lower Turning Points of Business Cycles – A Multivariate Linear Discriminant Analysis for Germany 1974 to 2009

  • Ullrich HeilemannEmail author
  • Heinz Josef Münch
Part of the Studies in Classification, Data Analysis, and Knowledge Organization book series (STUDIES CLASS)


This paper examines with multivariate linear discriminant analysis (LDA) the accuracy of dating lower turning point phases (LTP) of a four-phase classification scheme of German business cycles from 1974 to 2009, in particular why the start of the Great Recession (GR) is dated so late. Based on quarterly year-to-year rates of change, 85% of LTP periods are classified correctly, which is the lowest rate of the four phases but is not different from the results for previous periods. The misclassification of the start of the GR as an upper turning point phase (UTP) and thereby skipping the DOWN phase seems to be the result of a rare coincidence of outliers of influential classifiers. The re-estimation of the classification functions with quarter-to-quarter rates of change showed promising results; however, they also failed to classify the start of the GR correctly.


  1. Belsley, D. A., Kuh, E., & Welch, R. E. (1980). Regression diagnostics. New York, NY: Wiley.CrossRefGoogle Scholar
  2. Döhrn, R., Barabas, G., Gebhardt, H., Middendorf, T., Milton, A. R., Münch, H. J., et al. (2005). Die wirtschaftliche Entwicklung im Inland: Weiterhin kein kräftiger Aufschwung. RWI-Konjunkturberichte, 56, 23–58.Google Scholar
  3. Economic Cycle Research Institute (ECRI) (Ed.). (2010), Business cycle peak and trough Dates, 21 Countries, 1948–2009. Retrieved September 25, 2018, from
  4. Heilemann, U. (2019). Rezessionen in der Bundesrepublik Deutschland von 1966 bis 2013. In Wirtschaftsdienst, 99, 1–7.CrossRefGoogle Scholar
  5. Heilemann, U., & Münch, H. J. (1999). Classification of West German business cycles 1955–1994. Jahrbücher für Nationalökonomie und Statistik, 219, 632–656.CrossRefGoogle Scholar
  6. Heilemann, U., & Schnorr-Bäcker, S. (2017). Could the start of the German recession 2008–2009 have been foreseen? Evidence from real-time data. Jahrbücher für Nationalökonomie und Statistik, 237, 29–62.Google Scholar
  7. Heilemann, U., & Weihs, C. (2003). Ergebnisse Teilprojekt B3 “Multivariate Bestimmung und Untersuchung von Konjunkturzyklen”. In Sonderforschungsbereich 475 “Komplexitätsreduktion in multivariaten Datenstrukturen” an der Universität Dortmund (Hrsg.), Dortmund (pp. 135–156).Google Scholar
  8. Layton, A. P., & Banerji, A. (2003). What is a recession?: A reprise. Applied Economics, 35(16), 1789–1797. Scholar
  9. Reinhart, C. M., & Rogoff, K. S. (2009). This time is different—Eight centuries of financial folly. Princeton, NJ: Princeton University Press.CrossRefGoogle Scholar
  10. Röhl, M. C. (1998). Computerintensive Dimensionsreduktion in der Klassifikation. Dortmunder dissertation. Lohmar: Eul Verlag.Google Scholar
  11. Sachverständigenrat zur Begutachtung der gesamtwirtschaftlichen Entwicklung. (2017). Jahresgutachten 2017/18. Wiesbaden: Statistisches Bundesamt.Google Scholar
  12. Zarnowitz, V. (2006). Phases and stages of recent U.S. business cycles. In U. Heilemann & C. Weihs (Eds.), Classification and clustering in business cycle analysis (RWI Schriften, 79.) (pp. 19–32). Essen: RWI.Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Universität LeipzigLeipzigGermany
  2. 2.EssenGermany

Personalised recommendations