Estimating a Causal Order among Groups of Variables in Linear Models
By Doris Entner, Patrik O. Hoyer. 2012
18 citations recorded by the source corpus.
The machine learning community has recently devoted much attention to the problem of inferring causal relationships from statistical data. Most of this work has focused on uncovering connections among scalar random variables. We generalize existing methods to apply to collections of multi-dimensional random vectors, focusing on techniques applicable to linear models. The performance of the resulting algorithms is evaluated and compared in simulations, which show that our methods can, in many cases, provide useful information on causal relationships even for relatively small sample sizes.
Subfields: Artificial Intelligence, Statistics and Probability
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- David Lopez-Paz
- Krzysztof Chalupka
- Pietro Perona
- Frederick Eberhardt
- Bernhard Schölkopf
- Daniel Hernández-Lobato
- Pablo Morales-Mombiela
- Alberto Suárez
In Statistics and Probability
- Daniel Hernández-Lobato
- David Lopez-Paz
- Pablo Morales-Mombiela
- Alberto Suárez
- Pietro Perona
- Samuel Kaski
- Pekka Parviainen
- Frederick Eberhardt
Citations in the corpus
- Multi-Level Cause-Effect Systems (2015, background): “Finally, we note that our work is orthogonal to recent efforts to learn causal graphs over high-dimensional variables (Entner and Hoyer, 2012; Parviainen and Kaski, 2015).”
- Learning structures of Bayesian networks for variable groups (2015, background): “, Bayesian network for lossless dataset compression, (Proceedings of the Fifth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD, (1999)), 387-391 [9] Entner, D.”
- Bayesian Networks for Variable Groups (2015, background): “Entner and Hoyer (2012) have presented an algorithm for finding causal structures among groups of continuous variables.”
- Non-linear Causal Inference using Gaussianity Measures (2014, background): “LINGAM has been extended to consider linear relations among groups of variables (Entner and Hoyer, 2012; Kawahara et al., 2012). In real-world data, causal relationships tend to be non-linear, a fact that questions the usefulness of linear methods. Hoyer et al. (2009) show that a basic linear framew”