2016
DOI: 10.1111/rssb.12167
|View full text |Cite
|
Sign up to set email alerts
|

Causal Inference by using Invariant Prediction: Identification and Confidence Intervals

Abstract: Summary.What is the difference between a prediction that is made with a causal model and that with a non-causal model? Suppose that we intervene on the predictor variables or change the whole environment. The predictions from a causal model will in general work as well under interventions as for observational data. In contrast, predictions from a non-causal model can potentially be very wrong if we actively intervene on variables. Here, we propose to exploit this invariance of a prediction under a causal model… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

4
677
0

Year Published

2016
2016
2024
2024

Publication Types

Select...
7
1
1

Relationship

2
7

Authors

Journals

citations
Cited by 583 publications
(681 citation statements)
references
References 119 publications
(176 reference statements)
4
677
0
Order By: Relevance
“…There exist methods for causal discovery from changes of multiple data sets [Hoover, 1990; Tian and Pearl, 2001; Peters et al ., 2016] by exploiting the property of invariance of causal mechanisms. They used linear models to represent causal mechanism and, as a consequence, the invariance of causal mechanisms can be assessed by checking whether the involved parameters change across data sets or not.…”
Section: Cd-nod Phase 2: Nonstationarity Helps Determine Causal Dirmentioning
confidence: 99%
“…There exist methods for causal discovery from changes of multiple data sets [Hoover, 1990; Tian and Pearl, 2001; Peters et al ., 2016] by exploiting the property of invariance of causal mechanisms. They used linear models to represent causal mechanism and, as a consequence, the invariance of causal mechanisms can be assessed by checking whether the involved parameters change across data sets or not.…”
Section: Cd-nod Phase 2: Nonstationarity Helps Determine Causal Dirmentioning
confidence: 99%
“…We then have that SðEÞ% as E%, [6] meaning that for E 2 ⊇ E 1 we have SðE 2 Þ ⊇ SðE 1 Þ. Sufficient conditions under which SðEÞ = S* , that is the causal variables are uniquely identifiable, have been worked out for linear Gaussian SEMs, requiring that E is sufficiently "rich" and form a certain class of interventions (31). From [6], we conclude that with a larger amount of experimental settings ("more heterogeneity") we have higher degree of identifiability of causal effects.…”
Section: Causal Inference Based On Invariance Across Experimentsmentioning
confidence: 99%
“…To the best of our knowledge, however, the work in ref. 31 is the first of its kind that exploits invariance of conditional distributions for statistical estimation and confidence statements.…”
Section: Causal Inference Based On Invariance Across Experimentsmentioning
confidence: 99%
“…Other excluded methods that make use of interventional data include Cooper and Yoo (1999); Tian and Pearl (2001) and Eaton and Murphy (2007), where the latter does not require knowledge of the precise location of interventions in a similar spirit to Rothenhäusler et al (2015). Hyttinen et al (2012) also makes use of intervention data to learn feedback models, assuming do-interventions, while Peters et al (2016) permits to build a graph nodewise by estimating the parental set of each node separately.…”
Section: Considered Methodsmentioning
confidence: 99%