Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2022
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2209.08411 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917596210659328 |
|---|---|
| author | Liu, Siqi Lehrmann, Andreas |
| author_facet | Liu, Siqi Lehrmann, Andreas |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2209_08411 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | DynaConF: Dynamic Forecasting of Non-Stationary Time Series Liu, Siqi Lehrmann, Andreas Machine Learning 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. |
| title | DynaConF: Dynamic Forecasting of Non-Stationary Time Series |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2209.08411 |