LTSM-Bundle: A Toolbox and Benchmark on Large Language Models for Time Series Forecasting

Time Series Forecasting (TSF) has long been a challenge in time series analysis. Inspired by the success of Large Language Models (LLMs), researchers are now developing Large Time Series Models (LTSMs), universal transformer-based models that use autoregressive prediction to improve TSF. However, training LTSMs on heterogeneous time series data poses unique challenges, including diverse frequencies, dimensions, scalability, and patterns across datasets. Though recent efforts have studied and evaluated various design choices aimed at enhancing LTSM training and generalization capabilities, these design choices are typically studied and evaluated in isolation and are not compared collectively. In this work, we introduce LTSM-Bundle, a comprehensive toolbox and benchmark for training LTSMs, spanning pre-processing techniques, model configurations, and dataset configurations. Modularized and benchmarked LTSMs from multiple dimensions, encompassing prompting strategies, tokenization approaches, training paradigms, base model selection, data quantity, and dataset diversity. Our findings provide practical guidance for configuring effective LTSMs in real-world settings.