Learning Directed Acyclic Graphs with Hidden Variables via Latent Gaussian Graphical Model Selection
Authors:
- Benjamin FrotBFBenjamin Frot
- Preetam NandyPNPreetam Nandy
- Marloes H. MaathuisMarloes H. Maathuis
Preprint
Abstract
We introduce a new method to estimate the Markov equivalence class of a directed acyclic graph (DAG) in the presence of hidden variables, in settings where the underlying DAG among the observed variables is sparse, and there are a few hidden variables that have a direct effect on many of the observed ones. Building on the so-called low rank plus sparse framework, we suggest a two-stage approach which first removes unwanted variation using latent Gaussian graphical model selection, and then estimates the Markov equivalence class of the underlying DAG by applying GES. This approach is consistent in certain high-dimensional regimes and performs favourably when compared to the state of the art, both in terms of graphical structure recovery and total causal effect estimation.Keywords
All Related Versions
This publication has 22 references indexed in Scilit:
- Estimation of Directed Acyclic Graphs Through Two-Stage Adaptive Lasso for Gene Network InferenceJournal of the American Statistical Association, 2016
- TRRUST: a reference database of human transcriptional regulatory interactionsScientific Reports, 2015
- Alternating Direction Methods for Latent Variable Gaussian Graphical Model SelectionNeural Computation, 2013
- Using probabilistic estimation of expression residuals (PEER) to obtain increased power and interpretability of gene expression analysesNature Protocols, 2012
- Predicting causal effects in large-scale systems from observational dataNature Methods, 2010
- Markov equivalence for ancestral graphsThe Annals of Statistics, 2009
- On the completeness of orientation rules for causal discovery in the presence of latent confounders and selection biasArtificial Intelligence, 2008
- The Adaptive Lasso and Its Oracle PropertiesJournal of the American Statistical Association, 2006
- The max-min hill-climbing Bayesian network structure learning algorithmMachine Learning, 2006
- Transcription factors as targets for cancer therapyNature Reviews Cancer, 2002