Conformal prediction

We consider the problem of constructing distribution-free prediction intervals for multi-step time series forecasting, with a focus on the temporal dependencies inherent in multi-step forecast errors. We establish that the optimal h-step-ahead …

Conformal prediction constitutes a flexible family of methods for predictive uncertainty quantification, which has recently drawn significant attention in the statistical literature because of its distribution-free, finite-sample, and model-agnostic …