Researchers have developed a new method for risk-sensitive estimation using Wasserstein distributionally robust optimization. This approach measures estimator performance based on the conditional value-at-risk (CVaR) of estimation errors. The method allows for the computation of optimal affine estimators by solving a semidefinite program, and has shown improved performance in electricity price forecasting tasks. AI
Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →
RANK_REASON This is a research paper detailing a new estimation method and its evaluation on a specific task.