Conformal Prediction for Time-series Forecasting with Change Points
By: Sophia Sun, Rose Yu
Potential Business Impact:
Helps computers guess future numbers even when they change.
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.
Similar Papers
Conformal Inference for Time Series over Graphs
Machine Learning (CS)
Helps computers guess better about changing data.
A Gentle Introduction to Conformal Time Series Forecasting
Methodology
Makes predictions more trustworthy even with changing data.
Conformal Prediction Meets Long-tail Classification
Machine Learning (CS)
Makes AI more sure about rare things it sees.