解决时间序列突变点下的不确定性量化问题，提升预测可靠性。
Conformal Prediction for Time-series Forecasting with Change Points
- 引入状态预测模型与在线校准结合，应对数据分布突变
- 在6个真实和合成数据集上验证有效性与自适应性提升
- 适合需要可靠置信区间的时间序列预测场景
置信预测被广泛视为时间序列不确定性量化的一种通用高效方法。然而，现有方法难以处理存在突变点（即数据生成过程突然变化）的时间序列。本文提出一种针对带突变点的时间序列的置信预测方法（CPTC），通过集成状态预测模型与在线置信预测，实现对非平稳时间序列不确定性的建模。我们在最弱假设下证明了CPTC的有效性与增强的自适应性，并在6个合成及真实世界数据集上展示了其优于现有先进基线的方法，表现出更高的有效性和自适应能力。
原文摘要 · Abstract (English)
Conformal prediction has been explored as a general and efficient way to provide uncertainty quantification for time series. However, current methods struggle to handle time series data with change points - sudden shifts in the underlying data-generating process. In this paper, we propose a novel Conformal Prediction for Time-series with Change points (CPTC) algorithm, addressing this gap by integrating a model to predict the underlying state with online conformal prediction to model uncertainties in non-stationary time series. We prove CPTC's validity and improved adaptivity in the time series setting under minimum assumptions, and demonstrate CPTC's practical effectiveness on 6 synthetic and real-world datasets, showing improved validity and adaptivity compared to state-of-the-art baselines.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。