Recovering from Selection Bias in Causal and Statistical Inference

@article{Bareinboim2014RecoveringFS,
  title={Recovering from Selection Bias in Causal and Statistical Inference},
  author={Elias Bareinboim and Jin Tian and Judea Pearl},
  journal={Probabilistic and Causal Inference},
  year={2014}
}
Selection bias is caused by preferential exclusion of units from the samples and represents a major obstacle to valid causal and statistical inferences; it cannot be removed by randomized experiments and can rarely be detected in either experimental or observational studies. In this paper, we provide complete graphical and algorithmic conditions for recovering conditional probabilities from selection biased data. We also provide graphical conditions for recoverability when unbiased data is… 
View on ACM
cs.purdue.edu
112 Citations
Highly Influential Citations
10
Background Citations
59
Methods Citations
18
Results Citations
8

Figures from this paper

112 Citations

Citation Type

Citation Type

Recovering Causal Effects from Selection Bias
TLDR
This paper provides graphical and algorithmic conditions for recov-erability of interventional distributions for when selection and confounding biases are both present and completely characterizes the class of causal effects that are recoverable in Markovian models, and is sufficient for Semi-Markovian Models.
Identification of Causal Effects in the Presence of Selection Bias
TLDR
This paper investigates the problem of identifiability of causal effects when both confounding and selection biases are simultaneously present, and proposes a new algorithm that subsumes the current state-of-the-art method based on the back-door criterion.
Generalized Adjustment Under Confounding and Selection Biases
TLDR
This work generalizes the notion of backdoor adjustment to account for both biases and leverage external data that may be available without selection bias to predict how the population will react when it undergoes a change (intervention), following a new, compulsory decision protocol.
Causal Effect Identification by Adjustment under Confounding and Selection Biases
TLDR
This paper derives a sufficient and necessary condition for recovering causal effects using covariate adjustment from an observational distribution collected under preferential selection and presents a complete algorithm with polynomial delay to find all sets of admissible covariates for adjustment when confounding and selection biases are simultaneously present and unbiased data is available.
Adjustment Criteria for Recovering Causal Effects from Missing Data
TLDR
This paper introduces a covariate adjustment formulation for controlling confounding bias in the presence of missing-not-at-random data and develops a necessary and sufficient condition for recovering causal effects using the adjustment.
Recovering from Selection Bias using Marginal Structure in Discrete Models
TLDR
It is shown that generic identifiability of causal effects is possible in a much wider class of causal models than had previously been known.
Selection Mechanisms and Their Consequences: Understanding and Addressing Selection Bias
TLDR
This work used causal directed acyclic graphs to demonstrate the types of bias that can result when selection into a sample is associated with the exposure or outcome of interest, or with both.
Causal Calculus in the Presence of Cycles, Latent Confounders and Selection Bias
We prove the main rules of causal calculus (also called do-calculus) for i/o structural causal models (ioSCMs), a generalization of a recently proposed general class of non-/linear structural causal…
Causal Calculus in the Presence of Cycles, Latent Confounders and Selection Bias
We prove the main rules of causal calculus (also called do-calculus) for i/o structural causal models (ioSCMs), a generalization of a recently proposed general class of non-/linear structural causal…
Instrumental Variables with Treatment-induced Selection: Exact Bias Results
Instrumental variables (IV) estimation suffers selection bias when the analysis conditions on the treatment. Judea Pearl's early graphical definition of instrumental variables explicitly prohibited…
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 42 REFERENCES
Controlling Selection Bias in Causal Inference
TLDR
This paper highlights several graphical and algebraic methods capable of mitigating and sometimes eliminating selection bias, and generalize and improve previously reported results, and identify the type of knowledge that need to be available for reasoning in the presence of selection bias.
On recovering a population covariance matrix in the presence of selection bias
This paper considers the problem of using observational data in the presence of selection bias to identify causal effects in the framework of linear structural equation models. We propose a criterion…
Graphical models for inference under outcome-dependent sampling
TLDR
This work considers situations where data have been collected such that the sampling depends on the outcome of interest and possibly fur- ther covariates, as for instance in case-control studies, and gives sufficient graphical conditions for testing and estimating the causal effect of exposure on outcome.
On the completeness of orientation rules for causal discovery in the presence of latent confounders and selection bias
Causal diagrams for empirical research
SUMMARY The primary aim of this paper is to show how graphical models can be used as a mathematical language for integrating statistical and subject-matter information. In particular, the paper…
Conditional Independence in Sample Selection Models
Adjustment Criteria in Causal Diagrams: An Algorithmic Perspective
TLDR
This work proves equivalences between existing as well as new criteria for adjustment and provides a new simplified but still equivalent notion of d-separation, enabling users to assess the effects of covariate adjustment on diagrams with tens to hundreds of variables interactively in real time.
Learning and evaluating classifiers under sample selection bias
TLDR
This paper formalizes the sample selection bias problem in machine learning terms and study analytically and experimentally how a number of well-known classifier learning methods are affected by it.
Linear Models: A Useful “Microscope” for Causal Analysis
TLDR
Basic techniques of linear path analysis are reviewed and it is demonstrated, using simple examples, how causal phenomena of non-trivial character can be understood, exemplified and analyzed using diagrams and a few algebraic steps.
Meta-Transportability of Causal Effects: A Formal Approach
TLDR
This paper provides a complete algorithm for computing the transport formula, that is, a way of fusing observational and experimental information to synthesize an unbiased estimate of the desired eects in the target domain.
...
1
2
3
4
5
...