TimeMar: Multi-Scale Autoregressive Modeling for Unconditional Time Series Generation

TL;DR

TimeMar introduces a multi-scale, structure-disentangled autoregressive framework for generating high-quality, long-term time series data by capturing hierarchical dependencies and separating trend and seasonal components.

Contribution

It proposes a novel multiscale generation framework with a dual-path VQ-VAE for disentangling components, improving quality and efficiency over existing methods.

Findings

Outperforms existing methods in quality on six datasets

Achieves high performance with fewer parameters

Excels in long-term sequence generation

Abstract

Generative modeling offers a promising solution to data scarcity and privacy challenges in time series analysis. However, the structural complexity of time series, characterized by multi-scale temporal patterns and heterogeneous components, remains insufficiently addressed. In this work, we propose a structure-disentangled multiscale generation framework for time series. Our approach encodes sequences into discrete tokens at multiple temporal resolutions and performs autoregressive generation in a coarse-to-fine manner, thereby preserving hierarchical dependencies. To tackle structural heterogeneity, we introduce a dual-path VQ-VAE that disentangles trend and seasonal components, enabling the learning of semantically consistent latent representations. Additionally, we present a guidance-based reconstruction strategy, where coarse seasonal signals are utilized as priors to guide the reconstruction of fine-grained seasonal patterns. Experiments on six datasets show that our approach produces higher-quality time series than existing methods. Notably, our model achieves strong performance with a significantly reduced parameter count and exhibits superior capability in generating high-quality long-term sequences. Our implementation is available at https://anonymous.4open.science/r/TimeMAR-BC5B.