Invariance of Clustering Operations in Causal Effect Identification
Abstract
Clustering variables in causal graphs reduces the size of the graph and simplifies causal inference. However, arbitrary clustering can alter crucial causal relations among variables and lead to erroneous conclusions. While the identifiability of a causal effect in the clustered graph implies the identifiability in the original graph under mild conditions, nonidentifiability in clustered graph does not imply nonidentifiability in the original graph without further assumptions. When both identifiability and nonidentifiability are preserved, the clustering operation is called identification invariant. We present a broad class of clustering operations that are identification invariant based on conditions related to the c-components of the original graph. Finally, we demonstrate use of the results in practical settings.
Related
- Hierarchical Clustering Can Jointly Satisfy Richness, Consistency, and Scale Invariance
- Partial Invariance for Classification under Structural Causal Models
- Epidemiological Causal Graph Identification: Challenges, Identifiability and Algorithms
- Symmetries and Causality: Causal Effect Identification Beyond IID Data
- Identification and Estimation of Causal Estimands with Missing Not at Random Data
BibTeX
@misc{nykanen2026invariance,
title = {{Invariance of Clustering Operations in Causal Effect Identification}},
author = {Jani Nykänen and Otto Tabell and Santtu Tikka and Juha Karvanen},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.03101},
eprint = {2610.03101},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.03101},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Invariance of Clustering Operations in Causal Effect Identification
AU - Jani Nykänen
AU - Otto Tabell
AU - Santtu Tikka
AU - Juha Karvanen
PY - 2026
DA - 2026/10/02
PB - arXiv
DO - 10.48550/arxiv.2610.03101
UR - https://arxiv.org/abs/2610.03101
AB - Clustering variables in causal graphs reduces the size of the graph and simplifies causal inference. However, arbitrary clustering can alter crucial causal relations among variables and lead to erroneous conclusions. While the identifiability of a causal effect in the clustered graph implies the identifiability in the original graph under mild conditions, nonidentifiability in clustered graph does not imply nonidentifiability in the original graph without further assumptions. When both identifiability and nonidentifiability are preserved, the clustering operation is called identification invariant. We present a broad class of clustering operations that are identification invariant based on conditions related to the c-components of the original graph. Finally, we demonstrate use of the results in practical settings.
ER -