Predictions in Quantile RegressionsReport as inadecuate

Predictions in Quantile Regressions - Download this document for free, or read online. Document in PDF available to download.

Two different tools to evaluate quantile regressionforecasts are proposed: MAD, to summarize forecast errors, and a fluctuationtest to evaluate in-sample predictions. The scores of the PISA test to evaluatestudents’ proficiency are considered. Growth analysis relates school attainmentto economic growth. The analysis is complemented by investigating the estimatedregression and predictions not only at the centre but also in the tails. Forout-of-sample forecasts, the estimates in one wave are employed to forecast thefollowing waves. The reliability of in-sample forecasts is controlled byexcluding the part of the sample selected by a specific rule: boys to predictgirls, public schools to forecast private ones, vocational schools to predictnon-vocational, etc. The gradient computed in the subset is compared to itsanalogue computed in the full sample in order to verify the validity of theestimated equation and thus of the in-sample predictions.


Predictions, Quantile Regressions, Gradient

Cite this paper

Furno, M. 2014 Predictions in Quantile Regressions. Open Journal of Statistics, 4, 504-517. doi: 10.4236-ojs.2014.47048.

Author: Marilena Furno



Related documents