Causal Modelling Combining Instantaneous and Lagged Effects: An Identifiable Model Based on Non-Gaussianity
Abstract
Causal analysis of continuous-valued variables typically uses either autoregressive models or linear Gaussian Bayesian networks with instantaneous effects. Estimation of Gaussian Bayesian networks poses serious identifiability problems, which is why it was recently proposed to use non-Gaussian models. Here, we show how to combine the non-Gaussian instantaneous model with autoregressive models. We show that such a non-Gaussian model is identifiable without prior knowledge of network structure, and we propose an estimation method shown to be consistent. This approach also points out how neglecting instantaneous effects can lead to completely wrong estimates of the autoregressive coefficients.
Cite
Text
Hyvärinen et al. "Causal Modelling Combining Instantaneous and Lagged Effects: An Identifiable Model Based on Non-Gaussianity." International Conference on Machine Learning, 2008. doi:10.1145/1390156.1390210Markdown
[Hyvärinen et al. "Causal Modelling Combining Instantaneous and Lagged Effects: An Identifiable Model Based on Non-Gaussianity." International Conference on Machine Learning, 2008.](https://mlanthology.org/icml/2008/hyvarinen2008icml-causal/) doi:10.1145/1390156.1390210BibTeX
@inproceedings{hyvarinen2008icml-causal,
title = {{Causal Modelling Combining Instantaneous and Lagged Effects: An Identifiable Model Based on Non-Gaussianity}},
author = {Hyvärinen, Aapo and Shimizu, Shohei and Hoyer, Patrik O.},
booktitle = {International Conference on Machine Learning},
year = {2008},
pages = {424-431},
doi = {10.1145/1390156.1390210},
url = {https://mlanthology.org/icml/2008/hyvarinen2008icml-causal/}
}