Research map / Time series / Causal and generative time series
SAMoSSA: Multivariate Singular Spectrum Analysis with Stochastic Autoregressive Noise
Abdullah Omar Alomar, Munther A. Dahleh, Sean Mann, Devavrat Shah
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 |
|---|---|---|---|
| add second-stage AR modeling on residuals to mSSA (ablation) | SAMoSSA · Traffic | mean R^2: 0.747 -> 0.776 | Improved |
| add second-stage AR modeling on residuals to mSSA (ablation) | SAMoSSA · Electricity | mean R^2: 0.605 -> 0.829 | Improved |
| add second-stage AR modeling on residuals to mSSA (ablation) | SAMoSSA · Exchange | mean R^2: 0.674 -> 0.731 | Improved |
| add second-stage AR modeling on residuals to mSSA (ablation) | SAMoSSA · Synthetic | mean R^2: 0.366 -> 0.476 | Improved |
| replace DeepAR with SAMoSSA | SAMoSSA · Traffic | mean R^2: 0.824 -> 0.776 | Hurt |
| replace DeepAR with SAMoSSA | SAMoSSA · Electricity | mean R^2: 0.764 -> 0.829 | Improved |
| replace ARIMA with SAMoSSA | SAMoSSA · Exchange | mean R^2: 0.756 -> 0.731 | Hurt |
| replace ARIMA with SAMoSSA | SAMoSSA · Electricity | mean R^2: <-10 -> 0.829 | Improved |
| replace LSTM with SAMoSSA | SAMoSSA · Exchange | mean R^2: -1.825 -> 0.731 | Improved |
| replace Prophet with SAMoSSA | SAMoSSA · Synthetic | mean R^2: -0.445 -> 0.476 | Improved |
| increase T from 200 to 500000 (more data) | SAMoSSA estimation · Synthetic harmonics+AR(1), N=10 | estimation error of f_1 and AR parameter | Improved |
| increase max AR root lambda* from 0.3 to 0.95 | SAMoSSA estimation · Synthetic harmonics+AR(1), N=10 | estimation error | Hurt |
Closest papers by meaning
- Chimera: Effectively Modeling Multivariate Time Series with 2-Dimensional State Space Models (NeurIPS 2024 · similarity 0.85)
- TimeEmb: A Lightweight Static-Dynamic Disentanglement Framework for Time Series Forecasting (NeurIPS 2025 · similarity 0.84)
- Multivariate Probabilistic Time Series Forecasting with Correlated Errors (NeurIPS 2024 · similarity 0.84)
- TimeES: Probabilistic and Deterministic Time Series Forecasting via Evolutionary Spectra (NeurIPS 2026 · similarity 0.84)
- Frequency Adaptive Normalization For Non-stationary Time Series Forecasting (NeurIPS 2024 · similarity 0.83)
- Revitalizing Multivariate Time Series Forecasting: Learnable Decomposition with Inter-Series Dependencies and Intra-Series Variations Modeling (ICML 2024 · similarity 0.83)
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.
