English

DynaConF: Dynamic Forecasting of Non-Stationary Time Series

Machine Learning 2024-02-27 v3 Machine Learning

Abstract

Deep learning has shown impressive results in a variety of time series forecasting tasks, where modeling the conditional distribution of the future given the past is the essence. However, when this conditional distribution is non-stationary, it poses challenges for these models to learn consistently and to predict accurately. In this work, we propose a new method to model non-stationary conditional distributions over time by clearly decoupling stationary conditional distribution modeling from non-stationary dynamics modeling. Our method is based on a Bayesian dynamic model that can adapt to conditional distribution changes and a deep conditional distribution model that handles multivariate time series using a factorized output space. Our experimental results on synthetic and real-world datasets show that our model can adapt to non-stationary time series better than state-of-the-art deep learning solutions.

Keywords

Cite

@article{arxiv.2209.08411,
  title  = {DynaConF: Dynamic Forecasting of Non-Stationary Time Series},
  author = {Siqi Liu and Andreas Lehrmann},
  journal= {arXiv preprint arXiv:2209.08411},
  year   = {2024}
}

Comments

Accepted by Transactions on Machine Learning Research (TMLR), 2024