RMISC: A Large-scale Real-world Multivariate Corpus for Time Series Foundation Models
For the time series foundation model community, this provides a high-quality real-world benchmark and demonstrates the value of real-world multivariate data over synthetic data for pretraining.
The authors created RMISC, a large-scale real-world multivariate time series corpus with ~200 datasets and 142 billion time points, and showed that pretraining TSFMs on it outperforms synthetic data, improving zero-shot generalization on in-distribution and out-of-distribution benchmarks.
Recent years have witnessed the emergence of multivariate modeling using time series foundation models (TSFMs), which achieve advanced zero-shot generalization. Modern multivariate TSFMs are predominantly pretrained on multivariate synthetic data, which is easier to scale but may fail to capture the complex temporal dynamics and cross-variable relationships present in real-world time series. This raises a key question: Whether and to what extent the leading TSFMs trained with the real-world corpus perform better than those trained with synthetic data? To answer this, we establish the RMISC corpus, a considerably large-scale, high-quality, openly accessible, real-world, and multivariate time series archive that contains around 200 datasets and 142 billion time points across diverse domains. Furthermore, we pretrain four advanced TSFMs on univariate, synthetic multivariate, and real-world multivariate data and evaluate their zero-shot generalization capabilities on standard in-distribution and out-of-distribution benchmarks. Experimental results show that incorporating real-world multivariate data predominantly improves the generalization performance for both univariate and multivariate TSFMs. These results provide a deeper understanding of how real-world multivariate data contributes to the development of stronger TSFMs.