Skip to main content
SpringerLink
Log in

Turning Points and Classification

  • Chapter
  • First Online:
Macroeconomic Forecasting in the Era of Big Data

Part of the book series: Advanced Studies in Theoretical and Applied Econometrics ((ASTA,volume 52))

Abstract

Economic time-series data is commonly categorized into a discrete number of persistent regimes. I survey a variety of approaches for real-time prediction of these regimes and the turning points between them, where these predictions are formed in a data-rich environment. I place particular emphasis on supervised machine learning classification techniques that are common to the statistical classification literature, but have only recently begun to be widely used in economics. I also survey Markov-switching models, which are often used for unsupervised classification of economic data. The approaches surveyed are computationally feasible when applied to large datasets, and the machine learning algorithms employ regularization and cross-validation to prevent overfitting in the face of many predictors. A subset of the approaches conduct model selection automatically in forming predictions. I present an application to real-time identification of US business cycle turning points based on a wide dataset of 136 macroeconomic and financial time series.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution
Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD 219.00
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 279.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info
Hardcover Book
USD 279.99
Price excludes VAT (USA)
  • Durable hardcover edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Notes

  1. 1.

    http://www.R-project.org/.

  2. 2.

    When converting \(\widehat {S}^c_{t+h}\left (X_t\right )\) into turning point predictions, one might also consider conversion rules that acknowledge class persistence. For example, multiple periods of elevated class probabilities could be required before a turning point into that class is predicted.

  3. 3.

    Generalizations of these metrics to the multi-class case generally proceed by considering each class against all other classes in order to mimic a two class problem.

  4. 4.

    In the classification literature, TPR is referred to as the sensitivity and TNR as the specificity.

  5. 5.

    In a CART classification tree, \(\mathcal {T}^{A,j}\) is a discrete set of all non-equivalent values for Ï„ j, which is simply the set of midpoints of the ordered values for X j,t in the training sample observations relevant for node A.

  6. 6.

    Other papers that were influential in the development of random forest methods include Amit and Geman (1997) and Ho (1998).

  7. 7.

    MS models generally require a normalization in order to properly define the regimes. For example, in a two regime example where the regimes are high and low volatility, we could specify that S t+h = 1 is the low variance regime and S t+h = 2 is the high variance regime. In practice this is enforced by restricting the variance in S t+h = 2 to be larger than that in S t+h = 1. See Hamilton, Waggoner, and Zha (2007) for an extensive discussion of normalization in the MS model.

  8. 8.

    See, e.g., Chauvet and Piger (2008), Chauvet and Hamilton (2006), and Giusto and Piger (2017).

  9. 9.

    In unreported results, I also considered a version of each supervised classifier that classified based on predictor variables formed as principal components from the relevant dataset. The performance of this version of the classifier was similar in all cases to the results applied to the full dataset of individual predictors.

  10. 10.

    This is a relatively small number of repeats, and was chosen to reduce the computational burden of the recursive out-of-sample nowcasting exercise. In unreported results, I confirmed the robustness of several randomly chosen reported results to a larger number of repeats (100).

References

  • Amit, Y., & Geman, D. (1997). Shape quantization and recognition with randomized trees. Neural Computation, 9, 1545–1588.

    Article  Google Scholar 

  • Ang, A., & Bekaert, G. (2002). Regime switches in interest rates. Journal of Business and Economic Statistics, 20, 163–182.

    Article  Google Scholar 

  • Berge, T. (2015). Predicting recessions with leading indicators: Model averaging and selection over the business cycle. Journal of Forecasting, 34(6), 455–471.

    Article  Google Scholar 

  • Berge, T., & Jordá, O. (2011). The classification of economic activity into expansions and recessions. American Economic Journal: Macroeconomics, 3(2), 246–277.

    Google Scholar 

  • Breiman, L. (1996). Bagging predictors. Machine Learning, 26(2), 123–140.

    Google Scholar 

  • Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32.

    Article  Google Scholar 

  • Breiman, L., Friedman, J., Olshen, R., & Stone, C. (1984). Classification and regression trees. Belmont, CA: Wadsworth.

    Google Scholar 

  • Camacho, M., Perez-Quiros, G., & Poncela, P. (2015). Extracting nonlinear signals from several economic indicators. Journal of Applied Econometrics, 30(7), 1073–1089.

    Article  Google Scholar 

  • Camacho, M., Perez-Quiros, G., & Poncela, P. (2018). Markov-switching dynamic factor models in real time. International Journal of Forecasting, 34, 598–611.

    Article  Google Scholar 

  • Chauvet, M. (1998). An econometric characterization of business cycle dynamics with factor structure and regime switching. International Economic Review, 39, 969–996.

    Article  Google Scholar 

  • Chauvet, M., & Hamilton, J. D. (2006). Dating business cycle turning points. In P. R. Costas Milas & D. van Dijk (Eds.), Nonlinear time series analysis of business cycles (pp. 1–53). North Holland: Elsevier.

    Google Scholar 

  • Chauvet, M., & Piger, J. (2008). A comparison of the real-time performance of business cycle dating methods. Journal of Business and Economic Statistics, 26(1), 42–49.

    Article  Google Scholar 

  • Cosslett, S., & Lee, L.-F. (1985). Serial correlation in discrete variable models. Journal of Econometrics, 27, 79–97.

    Article  Google Scholar 

  • D opke, J., Fritsche, U., & Pierdzioch, C. (2017). Predicting recessions with boosted regression trees. International Journal of Forecasting, 33, 745–759.

    Google Scholar 

  • Davig, T., & Smalter Hall, A. (2016). Recession forecasting using Bayesian classification. Federal Reserve Bank of Kansas City Research Working paper no. 1606.

    Google Scholar 

  • Diebold, F. X., & Rudebusch, G. D. (1996). Measuring business cycles: A modern perspective. The Review of Economics and Statistics, 78(1), 67–77.

    Article  Google Scholar 

  • Dueker, M. (1997). Markov switching in GARCH processes and mean-reverting stock market volatility. Journal of Business and Economic Statistics, 15, 26–34.

    Google Scholar 

  • Estrella, A., & Mishkin, F. S. (1998). Predicting U.S. recessions: Financial variables as leading indicators. The Review of Economics and Statistics, 80(1), 45–61.

    Article  Google Scholar 

  • Estrella, A., Rodrigues, A. P., & Schich, S. (2003). How stable is the predictive power of the yield curve? Evidence from Germany and the United States. The Review of Economics and Statistics, 85(3), 629–644.

    Article  Google Scholar 

  • Evans, M., & Wachtel, P. (1993). Inflation regimes and the sources of inflation uncertainty. Journal of Money, Credit and Banking, 25, 475–511.

    Article  Google Scholar 

  • Fossati, S. (2016). Dating U.S. business cycles with macro factors. Studies in Non-linear Dynamics and Econometrics, 20, 529–547.

    Google Scholar 

  • Freund, Y. (1995). Boosting a weak learning algorithm by majority. Information and Computation, 121(2), 256–285.

    Article  Google Scholar 

  • Freund, Y., & Schapire, R. (1996). Experiments with a new boosting algorithm. Proceedings of ICML, 13, 148–156.

    Google Scholar 

  • Friedman, J. (2001). Greedy function approximation: A gradient boosting machine. The Annals of Statistics, 29(5), 1189–1232.

    Article  Google Scholar 

  • Friedman, J., Hastie, T., & Tibshirani, R. (2000). Additive logistic regression: A statistical view of boosting. The Annals of Statistics, 28(2), 337–407.

    Article  Google Scholar 

  • Fushing, H., Chen, S.-C., Berge, T., & Jordá, O. (2010). A chronology of international business cycles through non-parametric decoding. SSRN Working Paper no: 1705758.

    Google Scholar 

  • Garbellano, J. (2016). Nowcasting recessions with machine learning: New tools for predicting the business cycle (Bachelor’s Thesis, University of Oregon).

    Google Scholar 

  • Garcia, R., & Perron, P. (1996). An analysis of the real interest rate under regime shifts. Review of Economics and Statistics, 78(1), 111–125.

    Article  Google Scholar 

  • Garcia, R., & Schaller, H. (2002). Are the effects of monetary policy asymmetric? Economic Inquiry, 40, 102–119.

    Article  Google Scholar 

  • Geurts, P., Ernst, D., & Wehenkel, L. (2006). Extremely randomized trees. Machine Learning, 63(1), 3–42.

    Article  Google Scholar 

  • Giusto, A., & Piger, J. (2017). Identifying business cycle turning points in real time with vector quantization. International Journal of Forecasting, 33, 174–184.

    Article  Google Scholar 

  • Goldfeld, S. M., & Quandt, R. E. (1973). A Markov model for switching regressions. Journal of Econometrics, 1(1), 3–16.

    Article  Google Scholar 

  • Guidolin, M., & Timmermann, A. (2005). Economic implications of bull and bear regimes in UK stock and bond returns. Economic Journal, 115(500), 111–143.

    Article  Google Scholar 

  • Hamilton, J. D. (1989). A new approach to the economic analysis of nonstationary time series and the business cycle. Econometrica, 57(2), 357–384.

    Article  Google Scholar 

  • Hamilton, J. D. (1994). Time series analysis. Princeton, NJ: Princeton University Press.

    Google Scholar 

  • Hamilton, J. D. (2008). Regime switching models. In S. N. Durlauf & L. E. Blume (Eds.), New Palgrave dictionary of economics (2nd ed.). London: Palgrave MacMillan.

    Google Scholar 

  • Hamilton, J. D. (2011). Calling recessions in real time. International Journal of Forecasting, 27(4), 1006–1026.

    Article  Google Scholar 

  • Hamilton, J. D., & Lin, G. (1996). Stock market volatility and the business cycle. Journal of Applied Econometrics, 11, 573–593.

    Article  Google Scholar 

  • Hamilton, J. D., & Raj, B. (2002). New directions in business cycle research and financial analysis. Empirical Economics, 27(2), 149–162.

    Article  Google Scholar 

  • Hamilton, J. D., & Susmel, R. (1994). Autoregressive conditional heteroskedasticity and changes in regime. Journal of Econometrics, 64, 307–333.

    Article  Google Scholar 

  • Hamilton, J. D., Waggoner, D. F., & Zha, T. (2007). Normalization in econometrics. Econometric Reviews, 26(2–4), 221–252.

    Article  Google Scholar 

  • Hammer, B., & Villmann, T. (2002). Generalized relevance learning vector quantization. Neural Networks, 15(8–9), 1059–1068.

    Article  Google Scholar 

  • Harding, D., & Pagan, A. (2006). Synchronization of cycles. Journal of Econometrics, 132(1), 59–79.

    Article  Google Scholar 

  • Hastie, T., Tibshirani, R., & Friedman, J. (2009). The elements of statistical learning. data mining, inference and prediction. New York, NY: Springer.

    Google Scholar 

  • Ho, T. K. (1998). The random subspace method for constructing decision forests. IEEE Transactions on Pattern Analysis and Machine Intelligence, 20(8), 832–844.

    Article  Google Scholar 

  • Karnizova, L., & Li, J. (2014). Economic policy uncertainty, financial markets and probability of U.S. recessions. Economics Letters, 125(2), 261–265.

    Article  Google Scholar 

  • Kaufmann, S. (2002). Is there an asymmetric effect of monetary policy over time? A Bayesian analysis using Austrian data. Empirical Economics, 27, 277–297.

    Article  Google Scholar 

  • Kauppi, H., & Saikkonen, P. (2008). Predicting U.S. recessions with dynamic binary response models. The Review of Economics and Statistics, 90(4), 777–791.

    Article  Google Scholar 

  • Kim, C.-J. (1994). Dynamic linear models with Markov switching. Journal of Econometrics, 60(1–2), 1–22.

    Article  Google Scholar 

  • Kim, C.-J. (2004). Markov-switching models with endogenous explanatory variables. Journal of Econometrics, 122, 127–136.

    Article  Google Scholar 

  • Kim, C.-J., Morley, J., & Piger, J. (2005). Nonlinearity and the permanent effects of recessions. Journal of Applied Econometrics, 20(2), 291–309.

    Article  Google Scholar 

  • Kim, C.-J., & Nelson, C. R. (1998). Business cycle turning points, a new coincident index, and tests of duration dependence based on a dynamic factor model with regime switching. Review of Economics and Statistics, 80(2), 188–201.

    Article  Google Scholar 

  • Kim, C.-J., & Nelson, C. R. (1999a). Friedman’s plucking model of business fluctuations: Tests and estimates of permanent and transitory components. Journal of Money, Credit and Banking, 31, 317–334.

    Article  Google Scholar 

  • Kim, C.-J., & Nelson, C. R. (1999b). Has the U.S. economy become more stable? A Bayesian approach based on a Markov-switching model of the business cycle. Review of Economics and Statistics, 81(4), 608–616.

    Article  Google Scholar 

  • Kim, C.-J., & Nelson, C. R. (1999c). State-space models with regime switching. Cambridge, MA: The MIT Press.

    Google Scholar 

  • Kohonen, T. (2001). Self-organizing maps. Berlin: Springer.

    Book  Google Scholar 

  • Lo, M. C., & Piger, J. (2005). Is the response of output to monetary policy asymmetric? Evidence from a regime-switching coefficients model. Journal of Money, Credit and Banking, 37, 865–887.

    Article  Google Scholar 

  • Manela, A., & Moreira, A. (2017). News implied volatility and disaster concerns. Journal of Financial Economics, 123(1), 137–162.

    Article  Google Scholar 

  • Mayr, A., Binder, H., Gefeller, O., & Schmid, M. (2014). The evolution of boosting algorithms: From machine learning to statistical modelling. Methods of Information in Medicine, 6(1), 419–427.

    Google Scholar 

  • McCracken, M. W., & Ng, S. (2015). Fred-md: A monthly database for macroeconomic research. St. Louis Federal Reserve Bank Working Paper no. 2015-012B.

    Google Scholar 

  • Ng, S. (2014). Boosting recessions. Canadian Journal of Economics, 47(1), 1–34.

    Article  Google Scholar 

  • Owyang, M. T., Piger, J., & Wall, H. J. (2015). Forecasting national recessions using state-level data. Journal of Money, Credit and Banking, 47(5), 847–866.

    Article  Google Scholar 

  • Piger, J. (2009). Econometrics: Models of regime changes. In R. A. Meyers (Ed.), Encyclopedia of complexity and system science (pp. 2744–2757). Berlin: Springer.

    Chapter  Google Scholar 

  • Qi, M. (2001). Predicting U.S. recessions with leading indicators via neural network models. International Journal of Forecasting, 17, 383–401.

    Article  Google Scholar 

  • Ravn, M., & Sola, M. (2004). Asymmetric effects of monetary policy in the united states. Federal Reserve Bank of St. Louis Review, 86, 41–60.

    Google Scholar 

  • Rudebusch, G., & Williams, J. (2009). Forecasting recessions: The puzzle of the enduring power of the yield curve. Journal of Business and Economic Statistics, 27(4), 492–503.

    Article  Google Scholar 

  • Sato, A., & Yamada, K. (1995). Generalized learning vector quantization. In D. T. G. Tesauro & T. Leen (Eds.), Advances in neural information processing systems (pp. 423–429). North Holland: Elsevier.

    Google Scholar 

  • Schapire, R. E. (1990). The strength of weak learnability. Machine Learning, 5, 197–227.

    Google Scholar 

  • Sims, C. A., & Zha, T. (2006). Were there regime switches in U.S. monetary policy? American Economic Review, 96(1), 54–81.

    Article  Google Scholar 

  • Stock, J. H., & Watson, M. W. (2002). Forecasting using principal components from a large number of predictors. Journal of the American Statistical Association, 97, 1167–1179.

    Article  Google Scholar 

  • Stock, J. H., & Watson, M. W. (2014). Estimating turning points using large data sets. Journal of Econometrics, 178(1), 368–381.

    Article  Google Scholar 

  • Turner, C. M., Startz, R., & Nelson, C. R. (1989). A Markov model of heteroskedasticity, risk, and learning in the stock market. Journal of Financial Economics, 25(1), 3–22.

    Article  Google Scholar 

  • Vishwakarma, K. (1994). Recognizing business cycle turning points by means of a neural network. Computational Economics, 7, 175–185.

    Article  Google Scholar 

  • Ward, F. (2017). Spotting the danger zone: Forecasting financial crises with classification tree ensembles and many predictors. Journal of Applied Econometrics, 32, 359–378.

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Department of Economics, University of Oregon, Eugene, OR, USA

    Jeremy Piger

Authors
  1. Jeremy Piger
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Jeremy Piger .

Editor information

Editors and Affiliations

  1. UHERO and Department of Economics, University of Hawaii at Manoa, Honolulu, HI, USA

    Peter Fuleky

Rights and permissions

Reprints and permissions

Copyright information

© 2020 Springer Nature Switzerland AG

About this chapter

Check for updates. Verify currency and authenticity via CrossMark

Cite this chapter

Piger, J. (2020). Turning Points and Classification. In: Fuleky, P. (eds) Macroeconomic Forecasting in the Era of Big Data. Advanced Studies in Theoretical and Applied Econometrics, vol 52. Springer, Cham. https://doi.org/10.1007/978-3-030-31150-6_18

Download citation

Publish with us

Policies and ethics

Search

Navigation