Causal modelling combining instantaneous and lagged effects: an identifiable model based on non-Gaussianity

By Aapo Hyvärinen, Shohei Shimizu, Patrik O. Hoyer. 2008

102 citations recorded by the source corpus.

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

In Statistics and Probability

Citations in the corpus

Invariant Causal Prediction for Sequential Data (2017, methodology): “Other methods that are able to identify instantaneous effects (e.g., Chu and Glymour 2008; Hyvärinen, Shimizu, and Hoyer 2008; Peters, Janzing, and Schölkopf 2013) often require nonlinearities or non-Gaussianity.”
Statistical analysis of coupled time series with Kernel Cross-Spectral Density operators (2013, methodology): “Following [13, 18], our independence test can further be combined to recent developments in kernel time series prediction techniques [20] to define general and reliable multivariate causal inference techniques.”
Causal Inference on Time Series using Restricted Structural Equation Models (2013, methodology): “TS-LiNGAM [Hyvärinen et al., 2008] is based on LiNGAM [Shimizu et al.”
Estimation of a Structural Vector Autoregression Model Using Non-Gaussianity (2010, methodology): “Preliminary results were published in Hyvärinen et al. (2008) and Zhang and Hyv̈arinen (2009).”
Causal Inference by Independent Component Analysis: Theory and Applications* (2013, unknown): “Adapting the procedure to the identification of the SVAR model, the resulting VAR-LiNGAM algorithm is provided in Algorithm 1.4 Further details and discussion can be found in Hyvärinen et al. (2008).”
Causal modeling and inference for electricity markets (2011, methodology): “Hyvärinen et al. (2008) proposed a combined definition of Granger causality: If at least one of the coefficients Bτ (j, i), τ ≥ 0, is significantly non-zero, variable i causes j.”
Causal Inference on Time Series using Structural Equation Models (2012, methodology): “Note that linear functions lead to the model of Hyv¨arinen et al. [2008] as a special case.”
Discovering Temporal Causal Relations from Subsampled Data (2015, methodology): “The two-step method first estimates the errors in the above VAR model and then apply independent component analysis (ICA) (Hyvärinen et al., 2001) on the estimated errors Hyvärinen et al. (2008).”
Causal Discovery with Attention-Based Convolutional Neural Networks (2019, background): “Neglecting instantaneous influences can lead to misleading interpretations [13].”
Identifiability and estimation of structural vector autoregressive models for subsampled and mixed-frequency time series. (2017, methodology): “Our work concretely connects the non-Gaussian subsampled VAR with independent innovations method [14] with the now extensive non-Gaussian SVAR framework [15, 16, 17, 18, 19] by proving identifiability of an SVAR model of order one under arbitrary subsampling.”