Surrogate Modeling for Explainable Predictive Time Series Corrections
This work addresses the need for explainability in predictive time-series corrections for users of forecasting models, though it appears incremental as it builds on existing surrogate modeling techniques.
The paper tackles the problem of making time-series forecasting corrections interpretable by introducing a local surrogate approach that fits a base model to data with predicted errors removed, allowing parameter differences to serve as explanations, and demonstrates its potential through illustrative examples.
We introduce a local surrogate approach for explainable time-series forecasting. An initially non-interpretable predictive model to improve the forecast of a classical time-series 'base model' is used. 'Explainability' of the correction is provided by fitting the base model again to the data from which the error prediction is removed (subtracted), yielding a difference in the model parameters which can be interpreted. We provide illustrative examples to demonstrate the potential of the method to discover and explain underlying patterns in the data.