counterf. beamer present 080707
TRANSCRIPT
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
Potential Outcomes, Counterfactuals and Structural
Modelling
Causal Approaches in the Social Sciences
Federica Russoa, Guillaume Wunschb and Michel Mouchartc
aPhilosophy, bDemography and cStatisticsUniversity of Louvain (UCLouvain - Belgium)
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 1
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
Causal Analysis in Social Sciences
In agreement with Pearl(2000), two approaches to causality:
the potential outcome/counterfactual framework aschampioned most notably by Donald Rubinthe causal/structural modelling framework à la Wright,Haavelmo, Blalock, and others (including Pearl himself)
Counterfactuals, Not a new idea:to check what would happen were the putative cause absent ratherthan present.
A (possible!) historical path:David Hume (1748)
⇒ David Lewis (1973, 2004)⇒ Donald Rubin (1974, ...)
⇒ ...Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 2
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
This paper
Rubin’s model: very influential but also severely criticised.
In this paper, we tackle two questions:
(i) Are all the criticisms addressed to the potential outcomemodel sound? If so,
(ii) Are counterfactual questions to be dismissed altogether?
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 3
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
Outline
1 Rubin’s potential outcome modelRubin’s definition of a causal effectEpistemological flaws
2 Lewis’counterfactuals
3 Structural modelling: a general frameworkThe meaning of “structural”
4 Discussion and Conclusion
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 4
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
Rubin’s definition of a causal effectEpistemological flaws
Outline
1 Rubin’s potential outcome modelRubin’s definition of a causal effectEpistemological flaws
2 Lewis’counterfactuals
3 Structural modelling: a general frameworkThe meaning of “structural”
4 Discussion and Conclusion
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 5
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
Rubin’s definition of a causal effectEpistemological flaws
Measure of causal effect
Example: treatment of a headache
E : taking 2 aspirins
C : drinking a glass of water (control)
Yj(·): outcome of the treatment for individual j
Measure of the causal effect for individual j : Yj(E ) − Yj(C )
Average causal effect for a population of N individuals:
1
N
∑
1≤j≤N
Yj(E ) − Yj(C )
“Potential outcome" means: impossible to observe on a sameindividual Yj(E ) and Yj(C )
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 6
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
Rubin’s definition of a causal effectEpistemological flaws
“Potential outcome" issue
Rubin shows that
randomization
1/2[Y1(E ) − Y2(C )] + 1/2[Y2(E ) − Y1(C )]
or
(perfect) matching
Y1(E ) − Y2(C ) = Y2(E ) − Y1(C )
allows us to by-pass the potential outcome issue when evaluatingan average causal effect.
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 7
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
Rubin’s definition of a causal effectEpistemological flaws
Problems
Two challenges, however:
Individual heterogeneity: how to deal with the fact thatdifferent individuals may experience different effects from asame cause?
Assignment issue: in many observational studies, there areselection biases in the assignment of treatments
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 8
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
Rubin’s definition of a causal effectEpistemological flaws
Problems raised by Rubin’s approach
Potential outcomes: a "Platonic heaven"?i.e. one of the two potential outcomes will never be observed
Attributes: Rubin’s approach cannot take them into accounti.e. is "manipulation" an intrinsic ingredient of the concept ofcausality?
Causes of Effects: not only effects of causesIs experimentation an essential element of the concept ofcausality?
The individual or the population?i.e. statistics as a methodology of learning by observing aheterogeneous population of reference?
Counterfactuals or Causal Modelling?That’s the question...
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 9
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
Outline
1 Rubin’s potential outcome modelRubin’s definition of a causal effectEpistemological flaws
2 Lewis’counterfactuals
3 Structural modelling: a general frameworkThe meaning of “structural”
4 Discussion and Conclusion
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 10
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
Counterfactuals
Counterfactuals:
Subjunctive conditional statements the antecedent of whichstates a contrary-to-fact situation
Example
Statement “Had Mr Jones taken an aspirin half an hour ago,
his headache would have gone now”Presuppositions:
Mr Jones did not take the aspirin
Mr Jones still has headache
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 11
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
Counterfactuals
Counterfactuals:
Subjunctive conditional statements the antecedent of whichstates a contrary-to-fact situation
Example
Statement “Had Mr Jones taken an aspirin half an hour ago,
his headache would have gone now”Presuppositions:
Mr Jones did not take the aspirin
Mr Jones still has headache
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 12
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
Morals to be drawn from Lewis and Rubin
Lewis’counterfactuals are single-case,i.e. they concern a specific causal relation taking place at acertain time and space
e.g. Mr Jones (not) taking the aspirin and (not) havingheadache
Lewis’counterfactuals are not generic
e.g. whether aspirin is an effective treatment
e.g. an individual randomly sampled from the population
would recover from headache, were she to take aspirin
Counterfactual questions in Rubin’s model
share the same idea of Lewis
namely: had the cause not been, the effect would not have
been either
are not single-case but generic
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 13
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
Morals to be drawn from Lewis and Rubin
Lewis’counterfactuals are single-case,i.e. they concern a specific causal relation taking place at acertain time and space
e.g. Mr Jones (not) taking the aspirin and (not) havingheadache
Lewis’counterfactuals are not generic
e.g. whether aspirin is an effective treatment
e.g. an individual randomly sampled from the population
would recover from headache, were she to take aspirin
Counterfactual questions in Rubin’s model
share the same idea of Lewis
namely: had the cause not been, the effect would not have
been either
are not single-case but generic
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 14
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
Morals to be drawn from Lewis and Rubin
Lewis’counterfactuals are single-case,i.e. they concern a specific causal relation taking place at acertain time and space
e.g. Mr Jones (not) taking the aspirin and (not) havingheadache
Lewis’counterfactuals are not generic
e.g. whether aspirin is an effective treatment
e.g. an individual randomly sampled from the population
would recover from headache, were she to take aspirin
Counterfactual questions in Rubin’s model
share the same idea of Lewis
namely: had the cause not been, the effect would not have
been either
are not single-case but generic
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 15
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
To sum up
Are all criticisms addressed to the potential outcome model sound?
YESIn particular, the potential outcome model is problematic forestablishing generic causal claims.
An alternative: structural modelling
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 16
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
The meaning of “structural”
Outline
1 Rubin’s potential outcome modelRubin’s definition of a causal effectEpistemological flaws
2 Lewis’counterfactuals
3 Structural modelling: a general frameworkThe meaning of “structural”
4 Discussion and Conclusion
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 17
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
The meaning of “structural”
Introduction
Structural Models: models uncovering the structure of the datagenerating process and providing a causal explanation.
This involves
developing a conceptual model out of background knowledge
and then translating it into an operational model taking intoaccount the available indicators
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 18
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
The meaning of “structural”
Introduction
Structural Models: models uncovering the structure of the datagenerating process and providing a causal explanation.
This involves
developing a conceptual model out of background knowledge
and then translating it into an operational model taking intoaccount the available indicators
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 19
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
The meaning of “structural”
Introduction
Ingredients:
background knowledge(making sense of present data on the basis of pastobservations)
marginal-conditional decomposition(“explaining” by decomposing a complex mechanism into“simpler” pieces)
stability - invariance(disentangling structural from incidental)
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 20
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
The meaning of “structural”
Recursive Decomposition
Let us decompose X into p components: Suppose that the pcomponents of X = (X1, X2, · · ·Xp)Suppose that the components of X have been ordered in such away that in the complete decomposition:
pX (x | ω) = pXp |X1,X2,···Xp−1(xp | x1, x2, · · · xp−1, θp|1,···p−1)
· pXp−1|X1,X2,···Xp−2(xp−1 | x1, x2, · · · xp−2, θp−1|1,···p−2)
· · · · pX2|X1(x2 | x1, θ2|1) · pX1
(x1 | θ1), (1)
each component of the right hand side may be consideredstructural with mutually independent parameters:
ω = (θp|1,···p−1, θp−1|1,···p−2 · · · , θ1) ∈ Θp|1,···p−1×Θp−1|1,···p−2 · · ·×Θ1
(2)Then: (1) and (2) characterize a completely recursivedecomposition.
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 21
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
The meaning of “structural”
Recursive Decomposition
Let us decompose X into p components: Suppose that the pcomponents of X = (X1, X2, · · ·Xp)Suppose that the components of X have been ordered in such away that in the complete decomposition:
pX (x | ω) = pXp |X1,X2,···Xp−1(xp | x1, x2, · · · xp−1, θp|1,···p−1)
· pXp−1|X1,X2,···Xp−2(xp−1 | x1, x2, · · · xp−2, θp−1|1,···p−2)
· · · · pX2|X1(x2 | x1, θ2|1) · pX1
(x1 | θ1), (1)
each component of the right hand side may be consideredstructural with mutually independent parameters:
ω = (θp|1,···p−1, θp−1|1,···p−2 · · · , θ1) ∈ Θp|1,···p−1×Θp−1|1,···p−2 · · ·×Θ1
(2)Then: (1) and (2) characterize a completely recursivedecomposition.
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 22
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
The meaning of “structural”
Recursive Decomposition and Causality
Such a recursive decomposition provides a causal explanation aslong as:
1 each component of (1) represents a distinct mechanism
2 in each component, the explanatory (conditioning) variablesrepresent causal factors.i.e. background knowledge often involves conditionalindependence properties implying to drop some of theconditioning variables
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 23
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
Outline
1 Rubin’s potential outcome modelRubin’s definition of a causal effectEpistemological flaws
2 Lewis’counterfactuals
3 Structural modelling: a general frameworkThe meaning of “structural”
4 Discussion and Conclusion
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 24
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
The State of the Art: up to now
Causal Analysis concerns:
measuring effects of causes
providing causal explanations.
Rubin’s potential outcome/counterfactual approach: attention tothe importance of correctly assigning the units
to the treatment in experimental situations
to the control groups in non-experimental situations,
in order to avoid biases resulting from self-selection into the groups.
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 25
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
BUT: Flaws
BUT, methodological and epistemological flaws, e.g. :
As the manipulability of the cause is required, thecounterfactual approach cannot take attributes such as genderor ethnicity into account.
the approach is furthermore not adapted to the study of thecauses of an effect, but only to the effects of the causes
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 26
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
This paper: an alternative to Rubin’s model
A general structural modelling framework based on :
a thorough inventory of background knowledge required for
selecting the reference population
constructing the conceptual model composed of the relevant
variables and the putative causal relations among them, to be
transformed into an operational model .
a marginal-conditional - or recursive- decomposition of themultivariate distribution, represented in most cases by adirected acyclic graph, provided that the decomposition isstructurally valid (i.e. stability and invariance)
a non-parametric approach, more precisely: coordinate-free, orσ-algebraic, i.e. the fundamental- or: intrinsic- structure of themodel does not depend on an arbitrary choice of coordinates.
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 27
Rubin’s potential outcome modelLewis’counterfactuals
Structural modelling: a general frameworkDiscussion and Conclusion
Concluding Remarks
1 At variance from Pearl’s or Heckmann’s causal modellingapproaches, our causal framework does not imply a particularclass of statistical models
2 Compared to Rubin’s potential outcome/counterfactualframework, the structural modelling framework does notrequire interventions or manipulation of causes, and can dealboth with the effects of causes and with the causes of an effect
3 Qualitative methods also fall under this approach
Federica Russoa, Guillaume Wunschb and Michel Mouchartc Conterfactuals 28