The Relevance of AWS Chronos: An Evaluation of Standard Methods for Time Series Forecasting with Limited Tuning

TL;DR

This paper systematically compares Chronos, a transformer-based time series forecasting framework, with traditional models like ARIMA and Prophet, highlighting its advantages in long-term predictions and limited tuning scenarios.

Contribution

It provides a comprehensive evaluation of Chronos against traditional methods across various contexts and demonstrates its robustness with limited tuning and longer prediction horizons.

Findings

Chronos outperforms traditional models in long-term forecasting.

Traditional models' accuracy degrades with increased context length.

Prediction performance varies systematically across user classes.

Abstract

A systematic comparison of Chronos, a transformer-based time series forecasting framework, against traditional approaches including ARIMA and Prophet. We evaluate these models across multiple time horizons and user categories, with a focus on the impact of historical context length. Our analysis reveals that while Chronos demonstrates superior performance for longer-term predictions and maintains accuracy with increased context, traditional models show significant degradation as context length increases. We find that prediction quality varies systematically between user classes, suggesting that underlying behavior patterns always influence model performance. This study provides a case for deploying Chronos in real-world applications where limited model tuning is feasible, especially in scenarios requiring longer prediction.