Conformal Prediction Algorithms for Time Series Forecasting: Methods and Benchmark
This work addresses the problem of reliable uncertainty quantification in time series forecasting for practitioners, but it is incremental as it reviews and benchmarks existing methods rather than introducing new ones.
This review tackles the challenge of applying conformal prediction to time series forecasting, where temporal dependencies violate exchangeability assumptions, by surveying and benchmarking algorithmic solutions that relax these assumptions or adapt to distribution shifts, highlighting computational efficiency and practical performance on real-world data.
Reliable uncertainty quantification is of critical importance in time series forecasting, yet traditional methods often rely on restrictive distributional assumptions. Conformal prediction (CP) has emerged as a promising distribution-free framework for generating prediction intervals with rigorous theoretical guarantees. However, applying CP to sequential data presents a primary challenge: the temporal dependencies inherent in time series fundamentally violate the core assumption of data exchangeability, upon which standard CP guarantees are built. This review critically examines the main categories of algorithmic solutions designed to address this conflict. We survey and benchmark methods that relax the exchangeability assumption, those that redefine the data unit to be a collection of independent time series, approaches that explicitly model the dynamics of the prediction residuals, and online learning algorithms that adapt to distribution shifts to maintain long-run coverage. By synthesizing these approaches, we highlight computational efficiency and practical performance on real-world data.