Semiparametric Inference for Dynamic Causal Effects from Observational Time Series
Abstract
In observational time series, statistical inference for dynamic causal effects of a one-time intervention across horizons is complicated by high-dimensional observed pre-treatment information, unmeasured confounding, and serial dependence. To address these challenges, we develop a semiparametric framework for inference from a single serially dependent time series, integrating debiased machine learning with instrumental variables through buffered block cross-fitting. Under geometric beta-mixing, we derive non-asymptotic bounds on estimation error, asymptotic normality at each fixed horizon, and feasible inference that accommodates serial dependence. We further show how learner-specific prediction guarantees under temporal dependence can be used to verify the nuisance-rate conditions required for orthogonal inference. In a monetary-policy application with 468 months and 1464 lagged FRED-MD controls, we show an instrumented policy tightening lowers housing starts at medium horizons, with sensitivity analyses that support the finding.
Related
- Whole-brain modeling of dynamic causal circuits in human cognition using amortized variational inference
- Treatment persistence drives estimator performance in longitudinal causal inference based on observational data: A simulation study
- Causal Inference from Self-controlled Case Series
- GenAI Powered Dynamic Causal Inference with Unstructured Data
- Proximal Causal Inference for Time Series from N-of-1Mobile Device Data
BibTeX
@misc{yu2026semiparametric,
title = {{Semiparametric Inference for Dynamic Causal Effects from Observational Time Series}},
author = {Shibo Yu and Yan Chen and Jin-Hong Du and Guodong Li},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.27531},
eprint = {2609.27531},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.27531},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Semiparametric Inference for Dynamic Causal Effects from Observational Time Series
AU - Shibo Yu
AU - Yan Chen
AU - Jin-Hong Du
AU - Guodong Li
PY - 2026
DA - 2026/09/23
PB - arXiv
DO - 10.48550/arxiv.2609.27531
UR - https://arxiv.org/abs/2609.27531
AB - In observational time series, statistical inference for dynamic causal effects of a one-time intervention across horizons is complicated by high-dimensional observed pre-treatment information, unmeasured confounding, and serial dependence. To address these challenges, we develop a semiparametric framework for inference from a single serially dependent time series, integrating debiased machine learning with instrumental variables through buffered block cross-fitting. Under geometric beta-mixing, we derive non-asymptotic bounds on estimation error, asymptotic normality at each fixed horizon, and feasible inference that accommodates serial dependence. We further show how learner-specific prediction guarantees under temporal dependence can be used to verify the nuisance-rate conditions required for orthogonal inference. In a monetary-policy application with 468 months and 1464 lagged FRED-MD controls, we show an instrumented policy tightening lowers housing starts at medium horizons, with sensitivity analyses that support the finding.
ER -