CAARL: In-Context Learning for Interpretable Co-Evolving Time Series Forecasting
For practitioners needing interpretable forecasts of interdependent time series, CAARL offers a novel approach that balances accuracy and explainability, though improvements over existing methods are not quantified.
CAARL introduces an interpretable LLM-based framework for forecasting co-evolving time series by decomposing them into autoregressive segments, constructing temporal dependency graphs, and serializing them into narratives for chain-of-thought reasoning. It achieves competitive accuracy with state-of-the-art methods while providing transparent reasoning traces.
In this paper we investigate forecasting coevolving time series that feature intricate dependencies and nonstationary dynamics by using an LLM Large Language Models approach We propose a novel modeling approach named ContextAware ARLLM CAARL that provides an interpretable framework to decode the contextual dynamics influencing changes in coevolving series CAARL decomposes time series into autoregressive segments constructs a temporal dependency graph and serializes this graph into a narrative to allow processing by LLM This design yields a chainofthoughtlike reasoning path where intermediate steps capture contextual dynamics and guide forecasts in a transparent manner By linking prediction to explicit reasoning traces CAARL enhances interpretability while maintaining accuracy Experiments on realworld datasets validate its effectiveness positioning CAARL as a competitive and interpretable alternative to stateoftheart forecasting methods