Abstract
Information about forecast uncertainty is vital for optimal decision making in many domains that use weather forecasts. However, it is not available in the immediate output of deterministic numerical weather prediction systems. In this paper, we investigate several learning methods to train and evaluate prediction interval models of weather forecasts. The uncertainty models of weather predictions are trained from a database of historical forecasts/observations. They are developed to investigate prediction intervals of weather forecasts using various quantile regression methods as well as cluster-based probabilistic forecasts using fuzzy methods. To compare and verify probabilistic forecasts, a novel score is developed that accounts for sampling variation effects on forecast verification statistics. The impact of various feature sets and model parameters in forecast uncertainty modeling is also investigated. The results show superior performance of the non-linear quantile regression models in comparison with clustering methods.
Highlights
Introduction and BackgroundDeterministic Numerical Weather Prediction (NWP) models provide expected values of weather attributes on a three-dimensional spatial grid at certain forecast horizons [1]
Results confirm the higher quality of Prediction Intervals (PI) forecasts estimated by quantile regression models compared to the cluster-based models
It is followed by Local Quantile Regression (LocQR), Non-Linear Quantile Regression (NLQR), Kernel Quantile Regression (KQR), and LQR
Summary
Deterministic Numerical Weather Prediction (NWP) models provide expected values of weather attributes on a three-dimensional spatial grid at certain forecast horizons [1]. A detailed comparative study on the quality of non-parametric probabilistic forecasts of wind power and their statistical performance is evaluated in [28] It compares the fuzzy clustering based methods with the quantile regression-based approach in modeling PIs. An improved version of this approach is introduced in [30] where fuzzy clustering and error distribution fitting are applied to model the uncertainty of NWP forecasts. It presents a unique empirical and comparative study that covers a range of different cluster-based probabilistic models and quantile regression methods for modeling PIs of temperature and wind forecasts It develops a new hybrid clustering-quantile regression approach for PI modeling and evaluates its accuracy and performance.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.