DistMatch: Adaptive Binning via Distribution Matching for Robust Sequential Conformal Prediction
For practitioners needing reliable uncertainty quantification in time series with distributional shifts, DistMatch offers a more robust approach without requiring explicit reweighting.
DistMatch addresses the challenge of non-exchangeability in sequential conformal prediction for time series by using a binning method based on the Kolmogorov-Smirnov statistic to partition residuals into approximately exchangeable leaves, enabling locally adaptive inference. Experiments show it outperforms existing methods.
Sequential conformal prediction (CP) provides valid uncertainty quantification under the assumption of residual exchangeability. However, this assumption is often violated in real-world time series due to temporal dependencies and distributional shifts. While recent methods attempt to approximate exchangeability through reweighting, identifying optimal weights remains an open challenge. To address this limitation, we propose DistMatch, a binning-based method that recursively partitions residuals within a binary tree using the Kolmogorov-Smirnov (KS) statistic. We theoretically show that this partitioning induces approximately exchangeable leaves, thereby avoiding the need for reweighting. By applying quantile regression with online updates within each leaf, DistMatch enables locally adaptive inference and improves robustness to distributional shifts. Extensive experiments demonstrate that DistMatch outperforms existing sequential CP methods.