Foundation models for time series forecasting: Application in conformal prediction

TL;DR

This paper demonstrates that foundation models significantly improve the reliability and stability of conformal prediction intervals in time series forecasting, especially when data is scarce.

Contribution

It introduces the application of time series foundation models to conformal prediction, showing their advantages over traditional models in data-limited scenarios.

Findings

TSFMs provide more reliable prediction intervals with limited data.

Calibration stability is enhanced using TSFMs due to more data being utilized.

Benefits of TSFMs increase as data availability decreases.

Abstract

The zero-shot capabilities of foundation models (FMs) for time series forecasting offer promising potentials in conformal prediction, as most of the available data can be allocated to calibration. This study compares the performance of Time Series Foundation Models (TSFMs) with traditional methods, including statistical models and gradient boosting, within a conformal prediction setting. Our findings highlight two key advantages of TSFMs. First, when the volume of data is limited, TSFMs provide more reliable conformalized prediction intervals than classic models, thanks to their superior predictive accuracy. Second, the calibration process is more stable because more data are used for calibration. Morever, the fewer data available, the more pronounced these benefits become, as classic models require a substantial amount of data for effective training. These results underscore the potential of foundation models in improving conformal prediction reliability in time series applications, particularly in data-constrained cases. All the code to reproduce the experiments is available.