Conformal Prediction for Time-series Forecasting with Change Points
It addresses uncertainty quantification for time-series data with sudden shifts, which is a domain-specific problem for forecasting applications, but the approach appears incremental as it builds on existing conformal prediction methods.
The paper tackled the problem of uncertainty quantification in time-series forecasting with change points, proposing the CPTC algorithm that integrates state prediction with online conformal prediction, and demonstrated improved validity and adaptivity on 6 datasets compared to state-of-the-art baselines.
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.