Feature Fitted Online Conformal Prediction for Deep Time Series Forecasting Model

TL;DR

This paper introduces a lightweight, feature-based conformal prediction method for deep time series forecasting that provides valid, shorter confidence intervals without retraining, leveraging pre-trained models and adaptive coverage control.

Contribution

It presents a novel conformal prediction approach that uses features from pre-trained models to efficiently generate reliable confidence intervals with theoretical guarantees.

Findings

Achieves asymptotic coverage convergence with error bounds

Produces tighter confidence intervals while maintaining coverage

Demonstrates effectiveness on 12 diverse datasets

Abstract

Time series forecasting is critical for many applications, where deep learning-based point prediction models have demonstrated strong performance. However, in practical scenarios, there is also a need to quantify predictive uncertainty through online confidence intervals. Existing confidence interval modeling approaches building upon these deep point prediction models suffer from key limitations: they either require costly retraining, fail to fully leverage the representational strengths of deep models, or lack theoretical guarantees. To address these gaps, we propose a lightweight conformal prediction method that provides valid coverage and shorter interval lengths without retraining. Our approach leverages features extracted from pre-trained point prediction models to fit a residual predictor and construct confidence intervals, further enhanced by an adaptive coverage control mechanism. Theoretically, we prove that our method achieves asymptotic coverage convergence, with error bounds dependent on the feature quality of the underlying point prediction model. Experiments on 12 datasets demonstrate that our method delivers tighter confidence intervals while maintaining desired coverage rates. Code, model and dataset in \href{https://github.com/xiannanhuang/FFDCI}{Github}