Research map / Time series / Causal and generative time series
Spectral Subsampling MCMC for Stationary Time Series
Robert Salomone, Matias Quiroz, Robert Kohn, Mattias Villani, Minh-Ngoc Tran
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use spectral subsampling MCMC with the Whittle likelihood instead of full-data MCMC | Stationary time-series models · Vancouver, Stockholm, simulated ARFIMA, and Bitcoin time series | relative computational time: up to two orders of magnitude | Improved |
| Use Taylor control variates in spectral subsampling MCMC | Stationary time-series models · All four examples | posterior accuracy | Improved |
| Use grouped rather than individual-frequency Taylor control variates (ablation) | Taylor control variates · All four examples | variance | Improved |
| Use coreset control variates instead of Taylor control variates (ablation) | Subsampling MCMC · Time-series examples | posterior accuracy | Hurt |
| Use coreset control variates when groups contain only 22 observations (ablation) | Vancouver temperature ARMA(2,3) · Vancouver temperatures | control-variate usefulness: 22 | Hurt |
Closest papers by meaning
- Scalable Metropolis-Hastings for Exact Bayesian Inference with Large Datasets (ICML 2019 · similarity 0.89)
- On Markov chain Monte Carlo methods for tall data (JMLR 2017 · similarity 0.88)
- Sequential Monte Carlo Learning for Time Series Structure Discovery (ICML 2023 · similarity 0.87)
- BayOTIDE: Bayesian Online Multivariate Time Series Imputation with Functional Decomposition (ICML 2024 · similarity 0.85)
- Multi-Marginal Couplings for Metropolis--Hastings (NeurIPS 2026 · similarity 0.85)
- Parallelizing MCMC Across the Sequence Length (NeurIPS 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
