Results 1  10
of
13
Controlling Selection Bias in Causal Inference
, 2012
"... Selection bias, caused by preferential exclusion of samples from the data, is a major obstacle to valid causal and statistical inferences; it cannot be removed by randomized experiments and can hardly be detected in either experimental or observational studies. This paper highlights several graphica ..."
Abstract

Cited by 9 (7 self)
 Add to MetaCart
Selection bias, caused by preferential exclusion of samples from the data, is a major obstacle to valid causal and statistical inferences; it cannot be removed by randomized experiments and can hardly be detected in either experimental or observational studies. This paper highlights several graphical and algebraic methods capable of mitigating and sometimes eliminating this bias. These nonparametric methods generalize previously reported results, and identify the type of knowledge that is needed for reasoning in the presence of selection bias. Specifically, we derive a general condition together with a procedure for deciding recoverability of the odds ratio (OR) from sbiased data. We show that recoverability is feasible if and only if our condition holds. We further offer a new method of controlling selection bias using instrumental variables that permits the recovery of other effect measures besides OR. 1
Trygve Haavelmo and the Emergence of Causal Calculus
, 2012
"... Haavelmo was the first to recognize the capacity of economic models to guide policies. This paper describes some of the barriers that Haavelmo’s ideas have had (and still have) to overcome, and lays out a logical framework for capturing the relationships between theory, data and policy questions. Th ..."
Abstract

Cited by 8 (2 self)
 Add to MetaCart
Haavelmo was the first to recognize the capacity of economic models to guide policies. This paper describes some of the barriers that Haavelmo’s ideas have had (and still have) to overcome, and lays out a logical framework for capturing the relationships between theory, data and policy questions. The mathematical tools that emerge from this framework now enable investigators to answer complex policy and counterfactual questions using embarrassingly simple routines, some by mere inspection of the model’s structure. Several such problems are illustrated by examples, including misspecification tests, identification, mediation and introspection. Finally, we observe that modern economists are largely unaware of the benefits that Haavelmo’s ideas bestow upon them and, as a result, econometric research has not fully utilized modern advances in causal analysis. 1
Transportability of Causal Effects: Completeness Results
, 2012
"... The study of transportability aims to identify conditions under which causal information learned from experiments can be reused in a different environment where only passive observations can be collected. The theory introduced in [Pearl and Bareinboim, 2011] (henceforth [PB, 2011]) defines formal co ..."
Abstract

Cited by 7 (4 self)
 Add to MetaCart
The study of transportability aims to identify conditions under which causal information learned from experiments can be reused in a different environment where only passive observations can be collected. The theory introduced in [Pearl and Bareinboim, 2011] (henceforth [PB, 2011]) defines formal conditions for such transfer but falls short of providing an effective procedure for deciding whether transportability is feasible for a given set of assumptions about differences between the source and target domains. This paper provides such procedure. It establishes a necessary and sufficient condition for deciding when causal effects in the target domain are estimable from both the statistical information available and the causal information transferred from the experiments. The paper further provides a complete algorithm for computing the transport formula, that is, a way of fusing experimental and observational information to synthesize an estimate of the desired causal relation.
Causal inference by surrogate experiments: zidentifiability
, 2012
"... We address the problem of estimating the effect of intervening on a set of variables X from experiments on a different set, Z, that is more accessible to manipulation. This problem, which we call zidentifiability, reduces to ordinary identifiability when Z = ∅ and, like the latter, can be given sy ..."
Abstract

Cited by 5 (2 self)
 Add to MetaCart
We address the problem of estimating the effect of intervening on a set of variables X from experiments on a different set, Z, that is more accessible to manipulation. This problem, which we call zidentifiability, reduces to ordinary identifiability when Z = ∅ and, like the latter, can be given syntactic characterization using the docalculus [Pearl, 1995; 2000]. We provide a graphical necessary and sufficient condition for zidentifiability for arbitrary sets X, Z, and Y (the outcomes). We further develop a complete algorithm for computing the causal effect of X on Y using information provided by experiments on Z. Finally, we use our results to prove completeness of docalculus relative to zidentifiability, a result that does not follow from completeness relative to ordinary identifiability.
Some Thoughts Concerning Transfer Learning, with Applications to Metaanalysis and Datasharing Estimation
, 2012
"... A deeply entrenched axiom in the theory of learning states that the more one learns the easier it is to learn. In other words, the more proficient one becomes in performing familiar tasks, the easier it is to learn new tasks. This phenomenon, long recognized by psychologists and educators, has also ..."
Abstract

Cited by 4 (3 self)
 Add to MetaCart
A deeply entrenched axiom in the theory of learning states that the more one learns the easier it is to learn. In other words, the more proficient one becomes in performing familiar tasks, the easier it is to learn new tasks. This phenomenon, long recognized by psychologists and educators, has also been demonstrated in machine learning, especially in selftaught
Causal Transportability with Limited Experiments
, 2013
"... We address the problem of transferring causal knowledge learned in one environment to another, potentially different environment, when only limited experiments may be conducted at the source. This generalizes the treatment of transportability introduced in [Pearl and Bareinboim, 2011; Bareinboim and ..."
Abstract

Cited by 2 (1 self)
 Add to MetaCart
We address the problem of transferring causal knowledge learned in one environment to another, potentially different environment, when only limited experiments may be conducted at the source. This generalizes the treatment of transportability introduced in [Pearl and Bareinboim, 2011; Bareinboim and Pearl, 2012b], which deals with transferring causal information when any experiment can be conducted at the source. Given that it is not always feasible to conduct certain controlled experiments, we consider the decision problem whether experiments on a selected subset Z of variables together with qualitative assumptions encoded in a diagram may render causal effects in the target environment computable from the available data. This problem, which we call ztransportability, reduces to ordinary transportability when Z is allinclusive, and, like the latter, can be given syntactic characterization using the docalculus [Pearl, 1995; 2000]. This paper establishes a necessary and sufficient condition for causal effects in the target domain to be estimable from both the nonexperimental information available and the limited experimental information transferred from the source. We further provides a complete algorithm for computing the transport formula, that is, a way of fusing experimental and observational information to synthesize an unbiased estimate of the desired causal relation.
mTransportability: Transportability of a Causal Effect from Multiple Environments
"... {shlee, honavar} at iastate.edu We study mtransportability, a generalization of transportability, which offers a license to use causal information elicited from experiments and observations in m ≥ 1 source environments to estimate a causal effect in a given target environment. We provide a novel ch ..."
Abstract

Cited by 1 (1 self)
 Add to MetaCart
{shlee, honavar} at iastate.edu We study mtransportability, a generalization of transportability, which offers a license to use causal information elicited from experiments and observations in m ≥ 1 source environments to estimate a causal effect in a given target environment. We provide a novel characterization of mtransportability that directly exploits the completeness of docalculus to obtain the necessary and sufficient conditions for mtransportability. We provide an algorithm for deciding mtransportability that determines whether a causal relation is mtransportable; and if it is, produces a transport formula, that is, a recipe for estimating the desired causal effect by combining experimental information from m source environments with observational information from the target environment. 1
Transfer in Reinforcement Learning via Markov
"... n � Training and future (test) data n � follow the same distribution, and n � are in same feature space ..."
Abstract
 Add to MetaCart
n � Training and future (test) data n � follow the same distribution, and n � are in same feature space
External Validity and Transportability: A Formal Approach
, 2011
"... We provide a formal definition of the notion of “transportability, ” or “external validity, ” as a license to transfer causal information from experimental studies to a different population in which only observational studies can be conducted. We introduce a formal representation called “selection d ..."
Abstract
 Add to MetaCart
We provide a formal definition of the notion of “transportability, ” or “external validity, ” as a license to transfer causal information from experimental studies to a different population in which only observational studies can be conducted. We introduce a formal representation called “selection diagrams ” for expressing differences and commonalities between populations of interest and, using this representation, we derive procedures for deciding whether causal effects in the target population can be inferred from experimental findings in a different population. When the answer is affirmative, the procedures identify the set of experimental and observational studies that need be conducted to license the transport. We further discuss how transportability analysis can guide the transfer of knowledge in nonexperimental learning to minimize remeasurement cost and improve prediction power.