Title :
Discussion of “Prediction Intervals for Short-Term Wind Farm Generation Forecasts” and “Combined Nonparametric Prediction Intervals for Wind Power Generation”
Author :
Pinson, Pierre ; Tastu, Julija
Author_Institution :
Dept. of Electr. Eng., Tech. Univ. of Denmark, Lyngby, Denmark
Abstract :
A new score for the evaluation of interval forecasts, the so-called coverage width-based criterion (CWC), was proposed and utilized.. This score has been used for the tuning (in-sample) and genuine evaluation (out-ofsample) of prediction intervals for various applications, e.g., electric load [1], electricity prices [2], general purpose prediction [3], and wind power generation [4], [5]. Indeed, two papers by the same authors appearing in the IEEE Transactions On Sustainable Energy employ that score and use it to conclude on the comparative quality of alternative approaches to interval forecasting of wind power generation.
Keywords :
load forecasting; wind power plants; CWC; IEEE transactions on sustainable energy; alternative approach; combined nonparametric prediction interval; comparative quality; coverage width-based criterion; electric load; electricity prices; general purpose prediction; in-sample prediction interval; interval forecasts evaluation score; out-of-sample prediction interval; short-term wind farm generation forecast; wind power generation; Probabilistic logic; Wind forecasting; Wind power generation;
Journal_Title :
Sustainable Energy, IEEE Transactions on
DOI :
10.1109/TSTE.2014.2323851