Prediction of Quantiles by Statistical Learning and Application to GDP Forecasting

  • Pierre Alquier
  • Xiaoyin Li
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7569)


In this paper, we tackle the problem of prediction and confidence intervals for time series using a statistical learning approach and quantile loss functions. In a first time, we show that the Gibbs estimator is able to predict as well as the best predictor in a given family for a wide set of loss functions. In particular, using the quantile loss function of [1], this allows to build confidence intervals. We apply these results to the problem of prediction and confidence regions for the French Gross Domestic Product (GDP) growth, with promising results.


Statistical learning theory time series quantile regression GDP forecasting PAC-Bayesian bounds oracle inequalities weak dependence confidence intervals business surveys 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Koenker, R., Bassett, G.J.: Regression quantiles. Econometrica 46, 33–50 (1978)MathSciNetzbMATHCrossRefGoogle Scholar
  2. 2.
    Hamilton, J.: Time Series Analysis. Princeton University Press (1994)Google Scholar
  3. 3.
    Brockwell, P., Davis, R.: Time Series: Theory and Methods, 2nd edn. Springer (2009)Google Scholar
  4. 4.
    Cesa-Bianchi, N., Lugosi, G.: Prediction, Learning, and Games. Cambridge University Press, New York (2006)zbMATHCrossRefGoogle Scholar
  5. 5.
    Stoltz, G.: Agrégation séquentielle de prédicteurs: méthodologie générale et applications à la prévision de la qualité de l’air et à celle de la consommation électrique. Journal de la SFDS 151(2), 66–106 (2010)MathSciNetGoogle Scholar
  6. 6.
    Modha, D.S., Masry, E.: Memory-universal prediction of stationary random processes. IEEE Transactions on Information Theory 44(1), 117–133 (1998)MathSciNetzbMATHCrossRefGoogle Scholar
  7. 7.
    Meir, R.: Nonparametric time series prediction through adaptive model selection. Machine Learning 39, 5–34 (2000)zbMATHCrossRefGoogle Scholar
  8. 8.
    Alquier, P., Wintenberger, O.: Model selection for weakly dependent time series forecasting. Bernoulli 18(3), 883–913 (2012)Google Scholar
  9. 9.
    Biau, G., Patra, B.: Sequential quantile prediction of time series. IEEE Transactions on Information Theory 57, 1664–1674 (2011)MathSciNetCrossRefGoogle Scholar
  10. 10.
    Dalalyan, A., Tsybakov, A.: Aggregation by exponential weighting, sharp PAC-Bayesian bounds and sparsity. Machine Learning 72, 39–61 (2008)CrossRefGoogle Scholar
  11. 11.
    Gerchinovitz, S.: Sparsity regret bounds for individual sequences in online linear regression. In: Proceedings of COLT 2011 (2011)Google Scholar
  12. 12.
    Catoni, O.: Statistical Learning Theory and Stochastic Optimization. Springer Lecture Notes in Mathematics (2004)Google Scholar
  13. 13.
    Catoni, O.: PAC-Bayesian Supervised Classification (The Thermodynamics of Statistical Learning). Lecture Notes-Monograph Series, vol. 56. IMS (2007)Google Scholar
  14. 14.
    Littlestone, N., Warmuth, M.: The weighted majority algorithm. Information and Computation 108, 212–261 (1994)MathSciNetzbMATHCrossRefGoogle Scholar
  15. 15.
    Vovk, V.G.: Aggregating strategies. In: Proceedings of the 3rd Annual Workshop on Computational Learning Theory (COLT), pp. 372–283 (1990)Google Scholar
  16. 16.
    Koenker, R.: Quantile Regression. Cambridge University Press, Cambridge (2005)zbMATHCrossRefGoogle Scholar
  17. 17.
    Belloni, A., Chernozhukov, V.: L1-penalized quantile regression in high-dimensional sparse models. The Annals of Statistics 39(1), 82–130 (2011)MathSciNetzbMATHCrossRefGoogle Scholar
  18. 18.
    Shawe-Taylor, J., Williamson, R.: A PAC analysis of a bayes estimator. In: Proceedings of the Tenth Annual Conference on Computational Learning Theory, COLT 1997, pp. 2–9. ACM (1997)Google Scholar
  19. 19.
    McAllester, D.A.: PAC-Bayesian model averaging. In: Procs. of of the 12th Annual Conf. on Computational Learning Theory, Santa Cruz, California (Electronic), pp. 164–170. ACM, New York (1999)Google Scholar
  20. 20.
    Alquier, P.: PAC-Bayesian bounds for randomized empirical risk minimizers. Mathematical Methods of Statistics 17(4), 279–304 (2008)MathSciNetzbMATHCrossRefGoogle Scholar
  21. 21.
    Audibert, J.Y.: PAC-Bayesian aggregation and multi-armed bandits. HDR Université Paris Est (2010)Google Scholar
  22. 22.
    Audibert, J.Y., Catoni, O.: Robust linear least squares regression. The Annals of Statistics 39(5), 2766–2794 (2011)MathSciNetzbMATHCrossRefGoogle Scholar
  23. 23.
    Dedecker, J., Doukhan, P., Lang, G., León, J.R., Louhichi, S., Prieur, C.: Weak Dependence, Examples and Applications. Lecture Notes in Statistics, vol. 190. Springer, Berlin (2007)Google Scholar
  24. 24.
    Doukhan, P.: Mixing. Lecture Notes in Statistics. Springer, New York (1994)zbMATHCrossRefGoogle Scholar
  25. 25.
    Catoni, O.: A PAC-Bayesian approach to adaptative classification. Preprint Laboratoire de Probabilités et Modèles Aléatoires (2003)Google Scholar
  26. 26.
    Devilliers, M.: Les enquêtes de conjoncture. In: Archives et Documents. Number 101, INSEE (1984)Google Scholar
  27. 27.
    Clavel, L., Minodier, C.: A monthly indicator of the french business climate. Documents de Travail de la DESE (2009)Google Scholar
  28. 28.
    Dubois, E., Michaux, E.: Étalonnages à l’aide d’enquêtes de conjoncture: de nouvaux résultats. In: Économie et Prévision. Number 172. INSEE (2006)Google Scholar
  29. 29.
    Biau, G., Biau, O., Rouvière, L.: Nonparametric forecasting of the manufacturing output growth with firm-level survey data. Journal of Business Cycle Measurement and Analysis 3, 317–332 (2008)CrossRefGoogle Scholar
  30. 30.
    Diebold, F.X., Tay, A.S., Wallis, K.F.: Evaluating density forecasts of inflation: the survey of professional forecasters. Discussion Paper No.48, ESRC Macroeconomic Modelling Bureau, University of Warwick and Working Paper No.6228, National Bureau of Economic Research, Cambridge, Mass. (1997)Google Scholar
  31. 31.
    Tay, A.S., Wallis, K.F.: Density forecasting: a survey. Journal of Forecasting 19, 235–254 (2000)CrossRefGoogle Scholar
  32. 32.
    Britton, E., Fisher, P., Whitley, J.: The inflation report projections: Understanding the fan chart. Bank of England Quarterly Bulletin 38(1), 30–37 (1998)Google Scholar
  33. 33.
    Cornec, M.: Constructing a conditional gdp fan chart with an application to french business survey data. In: 30th CIRET Conference, New York (2010)Google Scholar
  34. 34.
    Dowd, K.: The inflation fan charts: An evaluation. Greek Economic Review 23, 99–111 (2004)Google Scholar
  35. 35.
    Li, X.: Agrégation de prédicteurs appliquée à la conjoncture. Rapport de stage de M2 - Université Paris 6 - INSEE sous la direction de Matthieu Cornec (2010)Google Scholar
  36. 36.
    Casella, G., Robert, C.: Monte Carlo Statistical Methods, 2nd edn. Springer (2004)Google Scholar
  37. 37.
    R Development Core Team: R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna (2008)Google Scholar
  38. 38.
    Rio, E.: Ingalités de Hoeffding pour les fonctions lipschitziennes de suites dépendantes. Comptes Rendus de l’Académie des Sciences de Paris, Série I 330, 905–908 (2000)MathSciNetzbMATHGoogle Scholar
  39. 39.
    Wintenberger, O.: Deviation inequalities for sums of weakly dependent time series. Electronic Communications in Probability 15, 489–503 (2010)MathSciNetzbMATHGoogle Scholar
  40. 40.
    Seldin, Y., Laviolette, F., Cesa-Bianchi, N., Auer, P., Shawe-Taylor, J.: PAC-Bayesian inequalities for martingales. Preprint arXiv:1110.6886 (2011)Google Scholar
  41. 41.
    Kullback, S.: Information theory and statistics. Wiley, New York (1959)zbMATHGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Pierre Alquier
    • 1
    • 3
  • Xiaoyin Li
    • 2
  1. 1.LPMA (Université Paris 7)ParisFrance
  2. 2.Laboratoire de Mathématiques (Université de Cergy-Pontoise)UCP site Saint-MartinCergy-PontoiseFrance
  3. 3.CREST (ENSAE)France

Personalised recommendations