Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2509.22217 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912608326516736 |
|---|---|
| author | Yao, Jie Zhang, Kai Rose, Eric Valachovic, Edward |
| author_facet | Yao, Jie Zhang, Kai Rose, Eric Valachovic, Edward |
| contents | Time series with multiple periodically correlated components is a complex problem with comparatively limited prior research. Most existing time series models are designed to accommodate simple periodically correlated components and tend to be sensitive to over-parameterization and optimization issues and are also unable to model complex PC components patterns in a time series. Frequency separation techniques can be used to maintain the correlation structure of each specific PC component, whereas Bayesian techniques can combine new and existing prior information to update beliefs about these components. This study introduces a method to combine the frequency separation techniques and Bayesian techniques to forecast PC and MPC time series data in a two stage form which is expected to show the new method's suitability in modeling MPC components compared to classical methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_22217 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Bayesian approach to the PC component Yao, Jie Zhang, Kai Rose, Eric Valachovic, Edward Methodology Time series with multiple periodically correlated components is a complex problem with comparatively limited prior research. Most existing time series models are designed to accommodate simple periodically correlated components and tend to be sensitive to over-parameterization and optimization issues and are also unable to model complex PC components patterns in a time series. Frequency separation techniques can be used to maintain the correlation structure of each specific PC component, whereas Bayesian techniques can combine new and existing prior information to update beliefs about these components. This study introduces a method to combine the frequency separation techniques and Bayesian techniques to forecast PC and MPC time series data in a two stage form which is expected to show the new method's suitability in modeling MPC components compared to classical methods. |
| title | Bayesian approach to the PC component |
| topic | Methodology |
| url | https://arxiv.org/abs/2509.22217 |