Feature Importance for Time Series Data: Improving KernelSHAP

TL;DR

This paper advances feature importance methods for time series data by providing closed-form SHAP solutions, applying KernelSHAP, and introducing Time Consistent Shapley values to enhance interpretability and event detection.

Contribution

It introduces new closed-form solutions for SHAP values in time series models and adapts KernelSHAP for time series, including a novel approach for event detection.

Findings

Closed-form SHAP solutions for VARMAX models

Application of KernelSHAP to time series tasks

Development of Time Consistent Shapley values

Abstract

Feature importance techniques have enjoyed widespread attention in the explainable AI literature as a means of determining how trained machine learning models make their predictions. We consider Shapley value based approaches to feature importance, applied in the context of time series data. We present closed form solutions for the SHAP values of a number of time series models, including VARMAX. We also show how KernelSHAP can be applied to time series tasks, and how the feature importances that come from this technique can be combined to perform "event detection". Finally, we explore the use of Time Consistent Shapley values for feature importance.