A1365
Title: Nonparametric quantile regression interval predictions for seasonal trending time series
Authors: Ida Bauer - University of Passau (Germany)
Harry Haupt - University of Passau (Germany)
Joachim Schnurbus - University of Passau (Germany) [presenting]
Abstract: Prediction intervals are developed based on nonparametric quantile regression for time series exhibiting seasonal patterns, possibly nonlinear trends, and trend-season interactions. Several novel bandwidth selection methods are proposed that either relate to quantile-specific measures or interval-specific measures. The performance of the proposed predictors is analyzed in a Monte Carlo simulation and for an energy demand application.
