Research map / Probabilistic and scientific modeling / Causal discovery
Estimating the Causal Effect from Partially Observed Time Series
Akane Iseki, Yusuke Mukuta, Yoshitaka Ushiku, Tatsuya Harada
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 semi-BPCCA (exploiting non-paired samples) instead of GSPCCA | semi-BPCCA · Synthetic data, T=50, MCAR/MNAR missing | estimated causality measure TE X->Y | Improved |
| Use semi-BPCCA instead of GSPCCA at T=200 | semi-BPCCA · Synthetic data, T=200 | estimated causality measure | No clear effect |
| Change missing pattern from MCAR to MNAR | semi-BPCCA · Synthetic data | estimated causality measure | No clear effect |
| Use semi-BPCCA instead of GSPCCA with few missing samples (missing ratio 0 to 0.4, T=100) | semi-BPCCA · Synthetic data, T=100 | estimated causality measure | Improved |
| Increase dimension of y from 40 to 60 and 80 (fewer samples than dimensions) | semi-BPCCA · Synthetic data, T=100 | estimated TE X->Y | Improved |
| Use semi-BPCCA instead of GSPCCA on the 3rd and 4th eigenvalue subspaces | semi-BPCCA · Synthetic data eigenvalue analysis | TE per eigen-subspace | Improved |
| Use semi-BPCCA instead of GSPCCA at 40% missing | semi-BPCCA · Meteorological data | average arrow length (km/day): 1.28 × 103 -> 1.09 × 103 | Improved |
| Use semi-BPCCA instead of GSPCCA at 20% missing | semi-BPCCA · Meteorological data | average arrow length (km/day): 1.01 × 103 -> 1.04 × 103 | No clear effect |
| Use semi-BPCCA instead of GSPCCA at 30% missing pixels | semi-BPCCA · Crowd Segmentation video causal flow | qualitative motion flow | Improved |
Closest papers by meaning
- Causal Discovery in Semi-Stationary Time Series (NeurIPS 2023 · similarity 0.87)
- Discovering Mixtures of Structural Causal Models from Time Series Data (ICML 2024 · similarity 0.86)
- Identification of Partially Observed Linear Causal Models: Graphical Conditions for the Non-Gaussian and Heterogeneous Cases (NeurIPS 2021 · similarity 0.86)
- Identifying Macro Conditional Independencies and Macro Total Effects in Summary Causal Graphs with Latent Confounding (AAAI 2025 · similarity 0.85)
- Local Identifying Causal Relations in the Presence of Latent Variables (ICML 2025 · similarity 0.85)
- Learning Causal Relations from Subsampled Time Series with Two Time-Slices (ICML 2024 · 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.
