Quantifying Uncertainty with a Derivative Tracking SDE Model and Application to Wind Power Forecast Data

Event Start
Event End
Location
KAUST

Abstract

We develop a data-driven methodology based on parametric Itô's Stochastic Differential Equations (SDEs) to capture forecast errors' asymmetric dynamics, including the forecast's uncertainty at time zero. Our SDE framework features time-derivative tracking of the forecast, time-varying mean-reversion parameter, and an improved state-dependent diffusion term. Proofs of the existence, strong uniqueness, and boundedness of the SDE solutions are shown by imposing conditions on the time-varying mean-reversion parameter. We develop the structure of the drift term based on sound mathematical theory. A truncation procedure regularizes the prediction function to ensure that the trajectories do not reach the boundaries almost surely in a finite time. Inference based on approximate likelihood, constructed through the moment-matching technique in the original forecast error space and the Lamperti space, is performed through numerical optimization procedures.  Another novel contribution is the characterization of the uncertainty of the forecast at time zero, which turns out to be crucial in practice.  All the procedures are agnostic of the forecasting technology, and they enable comparisons between different forecast providers. We apply our SDE framework to model historical Uruguayan normalized wind-power production and forecast data between April and December 2019. We obtain sharp empirical confidence bands of wind-power production forecast errors for the best-selected model.

Brief Biography

Raul Tempone's research interests are in the mathematical foundation of computational science and engineering. More specifically, he has focused on a posteriori error approximation and related adaptive algorithms for numerical solutions of various differential equations, including ordinary differential equations, partial differential equations, and stochastic differential equations. He is also interested in the development and analysis of efficient numerical methods for optimal control, uncertainty quantification and bayesian model calibration, validation and optimal experimental design. The areas of application he considers include, among others, engineering, chemistry, biology, physics as well as social science and computational finance.

Contact Person