Prediction of quantiles by statistical learning and application to GDP forecasting

Abstract : 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 (also known as Exponentially Weighted aggregate) 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 Koenker and Bassett (1978), 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.
Complete list of metadatas

Cited literature [40 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-00671982
Contributor : Pierre Alquier <>
Submitted on : Monday, August 6, 2012 - 3:36:16 PM
Last modification on : Tuesday, May 14, 2019 - 11:00:07 AM
Long-term archiving on: Friday, December 16, 2016 - 4:45:25 AM

Files

quantile6.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00671982, version 3
  • ARXIV : 1202.4294

Citation

Pierre Alquier, Xiaoyin Li. Prediction of quantiles by statistical learning and application to GDP forecasting. The Fifteenth International Conference on Discovery Science (DS 2012), Oct 2012, Lyon, France. pp.22-36. ⟨hal-00671982v3⟩

Share

Metrics

Record views

727

Files downloads

443