rename IATE to AIE

docs/src/user_guide/estimands.md

@@ -132,14 +132,14 @@ For a general higher-order definition, please refer to [Higher-order interaction
 For two points interaction with both treatment and control levels ``0`` and ``1`` for ease of notation:
 
 ```math
-IATE_{0 \rightarrow 1, 0 \rightarrow 1}(P) = \mathbb{E}[Y|do(T_1=1, T_2=1)] - \mathbb{E}[Y|do(T_1=1, T_2=0)]  \\
+AIE_{0 \rightarrow 1, 0 \rightarrow 1}(P) = \mathbb{E}[Y|do(T_1=1, T_2=1)] - \mathbb{E}[Y|do(T_1=1, T_2=0)]  \\
 - \mathbb{E}[Y|do(T_1=0, T_2=1)] + \mathbb{E}[Y|do(T_1=0, T_2=0)] 
 ```
 
 - Statistical Estimand (via backdoor adjustment):
 
 ```math
-IATE_{0 \rightarrow 1, 0 \rightarrow 1}(P) = \mathbb{E}_{\textbf{W}}[\mathbb{E}[Y|T_1=1, T_2=1, \textbf{W}] - \mathbb{E}[Y|T_1=1, T_2=0, \textbf{W}]  \\
+AIE_{0 \rightarrow 1, 0 \rightarrow 1}(P) = \mathbb{E}_{\textbf{W}}[\mathbb{E}[Y|T_1=1, T_2=1, \textbf{W}] - \mathbb{E}[Y|T_1=1, T_2=0, \textbf{W}]  \\
 - \mathbb{E}[Y|T_1=0, T_2=1, \textbf{W}] + \mathbb{E}[Y|T_1=0, T_2=0, \textbf{W}]] 
 ```
 
@@ -148,7 +148,7 @@ IATE_{0 \rightarrow 1, 0 \rightarrow 1}(P) = \mathbb{E}_{\textbf{W}}[\mathbb{E}[
 A causal estimand is given by:
 
 ```@example estimands
-causalΨ = IATE(
+causalΨ = AIE(
     outcome=:Y, 
     treatment_values=(
         T₁=(case=1, control=0), 
@@ -166,7 +166,7 @@ statisticalΨ = identify(causalΨ, scm)
 or defined directly:
 
 ```@example estimands
-statisticalΨ = IATE(
+statisticalΨ = AIE(
     outcome=:Y, 
     treatment_values=(
         T₁=(case=1, control=0), 
@@ -178,7 +178,7 @@ statisticalΨ = IATE(
 
 - Factorial Treatments
 
-It is possible to generate a `JointEstimand` containing all linearly independent IATEs from a set of treatment values or from a dataset. For that purpose, use the `factorialEstimand` function.
+It is possible to generate a `JointEstimand` containing all linearly independent AIEs from a set of treatment values or from a dataset. For that purpose, use the `factorialEstimand` function.
 
 ## Joint And Composed Estimands
 

docs/src/user_guide/estimation.md

@@ -195,10 +195,10 @@ result₃
 nothing # hide
 ```
 
-This time only the model for `Y` is fitted again while reusing the models for `T₁` and `T₂`. Finally, let's see what happens if we estimate the `IATE` between `T₁` and `T₂`.
+This time only the model for `Y` is fitted again while reusing the models for `T₁` and `T₂`. Finally, let's see what happens if we estimate the `AIE` between `T₁` and `T₂`.
 
 ```@example estimation
-Ψ₄ = IATE(
+Ψ₄ = AIE(
     outcome=:Y, 
     treatment_values=(
         T₁=(case=true, control=false), 
@@ -224,7 +224,7 @@ As explained in [Joint And Composed Estimands](@ref), a joint estimand is simply
 More precisely, we would like to see if the left-hand side of this equation is equal to the right-hand side:
 
 ```math
-IATE_{T_1=0 \rightarrow 1, T_2=0 \rightarrow 1} = ATE_{T_1=0 \rightarrow 1, T_2=0 \rightarrow 1} - ATE_{T_1=0, T_2=0 \rightarrow 1} - ATE_{T_1=0 \rightarrow 1, T_2=0}
+AIE_{T_1=0 \rightarrow 1, T_2=0 \rightarrow 1} = ATE_{T_1=0 \rightarrow 1, T_2=0 \rightarrow 1} - ATE_{T_1=0, T_2=0 \rightarrow 1} - ATE_{T_1=0 \rightarrow 1, T_2=0}
 ```
 
 For that, we need to define a joint estimand of three components:

docs/src/walk_through.md

@@ -111,7 +111,7 @@ marginal_ate_t1 = ATE(
 - The Interaction Average Treatment Effect:
 
 ```@example walk-through
-iate = IATE(
+aie = AIE(
     outcome = :Y,
     treatment_values = (
         T₁=(case=1, control=0), 
@@ -125,7 +125,7 @@ iate = IATE(
 Identification is the process by which a Causal Estimand is turned into a Statistical Estimand, that is, a quantity we may estimate from data. This is done via the `identify` function which also takes in the ``SCM``:
 
 ```@example walk-through
-statistical_iate = identify(iate, scm)
+statistical_aie = identify(aie, scm)
 ```
 
 Alternatively, you can also directly define the statistical parameters (see [Estimands](@ref)).
@@ -149,7 +149,7 @@ Statistical Estimands can be estimated without a ``SCM``, let's use the One-Step
 
 ```@example walk-through
 ose = OSE()
-result, cache = ose(statistical_iate, dataset)
+result, cache = ose(statistical_aie, dataset)
 result
 ```
 

src/TMLE.jl

@@ -28,7 +28,7 @@ using AutoHashEquals
 # #############################################################################
 
 export SCM, StaticSCM, add_equations!, add_equation!, parents, vertices
-export CM, ATE, IATE
+export CM, ATE, AIE
 export AVAILABLE_ESTIMANDS
 export factorialEstimand, factorialEstimands
 export TMLEE, OSE, NAIVE

src/counterfactual_mean_based/estimands.jl

@@ -33,7 +33,7 @@ variables(estimand::CMRelevantFactors) =
 const ESTIMANDS_DOCS = Dict(
     :CM => (formula="``CM(Y, T=t) = E[Y|do(T=t)]``",),
     :ATE => (formula="``ATE(Y, T, case, control) = E[Y|do(T=case)] - E[Y|do(T=control)``",),
-    :IATE => (formula="``IATE = E[Y|do(T₁=1, T₂=1)] - E[Y|do(T₁=1, T₂=0)] - E[Y|do(T₁=0, T₂=1)] + E[Y|do(T₁=0, T₂=0)]``",)
+    :AIE => (formula="``AIE = E[Y|do(T₁=1, T₂=1)] - E[Y|do(T₁=1, T₂=0)] - E[Y|do(T₁=0, T₂=1)] + E[Y|do(T₁=0, T₂=0)]``",)
 )
 
 for (estimand, (formula,)) ∈ ESTIMANDS_DOCS
@@ -89,6 +89,8 @@ for (estimand, (formula,)) ∈ ESTIMANDS_DOCS
     eval(ex)
 end
 
+const IATE = AIE
+
 CausalCMCompositeEstimands = Union{(eval(Symbol(:Causal, x)) for x in keys(ESTIMANDS_DOCS))...}
 
 StatisticalCMCompositeEstimand = Union{(eval(Symbol(:Statistical, x)) for x in keys(ESTIMANDS_DOCS))...}
@@ -109,7 +111,7 @@ end
 
 ncases(counterfactual_values, treatments_cases) = sum(counterfactual_values .== treatments_cases)
 
-function indicator_fns(Ψ::StatisticalIATE)
+function indicator_fns(Ψ::StatisticalAIE)
     N = length(Ψ.treatment_values)
     key_vals = Pair[]
     treatments_cases = Tuple(case_control.case for case_control ∈ values(Ψ.treatment_values))
@@ -233,7 +235,7 @@ unique_treatment_values(dataset, colnames) =
 """
 Generated from transitive treatment switches to create independent estimands.
 """
-get_treatment_settings(::Union{typeof(ATE), typeof(IATE)}, treatments_unique_values)=
+get_treatment_settings(::Union{typeof(ATE), typeof(AIE)}, treatments_unique_values)=
     sort(OrderedDict(key => collect(zip(uniquevaluess[1:end-1], uniquevaluess[2:end])) for (key, uniquevaluess) in pairs(treatments_unique_values)))
 
 get_treatment_settings(::typeof(CM), treatments_unique_values) = sort(OrderedDict(pairs(treatments_unique_values)))
@@ -304,7 +306,7 @@ end
 
 """
     factorialEstimand(
-        constructor::Union{typeof(CM), typeof(ATE), typeof(IATE)},
+        constructor::Union{typeof(CM), typeof(ATE), typeof(AIE)},
         treatments, outcome; 
         confounders=nothing,
         dataset=nothing, 
@@ -314,9 +316,9 @@ end
         verbosity=1
     )
 
-Generates a factorial `JointEstimand` with components of type `constructor` (CM, ATE, IATE). 
+Generates a factorial `JointEstimand` with components of type `constructor` (CM, ATE, AIE). 
 
-For the ATE and the IATE, the generated components are restricted to the Cartesian Product of single treatment levels transitions.
+For the ATE and the AIE, the generated components are restricted to the Cartesian Product of single treatment levels transitions.
 For example, consider two treatment variables T₁ and T₂ each taking three possible values (0, 1, 2). 
 For each treatment variable, the single treatment levels transitions are defined by (0 → 1, 1 → 2). 
 Then, the Cartesian Product of these transitions is taken, resulting in a 2 x 2 = 4 dimensional joint estimand:
@@ -332,7 +334,7 @@ A `JointEstimand` with causal or statistical components.
 
 # Args
 
-- `constructor`: CM, ATE or IATE.
+- `constructor`: CM, ATE or AIE.
 - `treatments`: An AbstractDictionary/NamedTuple of treatment levels (e.g. `(T=(0, 1, 2),)`) or a treatment iterator, then a dataset must be provided to infer the levels from it.
 - `outcome`: The outcome variable.
 - `confounders=nothing`: The generated components will inherit these confounding variables. If `nothing`, causal estimands are generated.
@@ -353,7 +355,7 @@ factorialEstimand(ATE, (T₁ = (0, 1), T₂=(0, 1, 2)), :Y₁)
 - An Average Interaction Effect with statistical components:
 
 ```@example
-factorial(IATE, (T₁ = (0, 1, 2), T₂=(0, 1, 2)), :Y₁, confounders=[:W₁, :W₂])
+factorial(AIE, (T₁ = (0, 1, 2), T₂=(0, 1, 2)), :Y₁, confounders=[:W₁, :W₂])
 ```
 
 - With a dataset, the treatment levels can be infered and a positivity constraint enforced:
@@ -369,7 +371,7 @@ factorialEstimand(ATE, [:T₁, :T₂], :Y₁,
 
 """
 function factorialEstimand(
-    constructor::Union{typeof(CM), typeof(ATE), typeof(IATE)},
+    constructor::Union{typeof(CM), typeof(ATE), typeof(AIE)},
     treatments, outcome; 
     confounders=nothing,
     dataset=nothing, 
@@ -393,7 +395,7 @@ end
 
 """
 factorialEstimands(
-    constructor::Union{typeof(ATE), typeof(IATE)},
+    constructor::Union{typeof(ATE), typeof(AIE)},
     dataset, treatments, outcomes; 
     confounders=nothing, 
     outcome_extra_covariates=(),
@@ -404,7 +406,7 @@ factorialEstimands(
 Generates a `JointEstimand` for each outcome in `outcomes`. See `factorialEstimand`.
 """
 function factorialEstimands(
-    constructor::Union{typeof(CM), typeof(ATE), typeof(IATE)},
+    constructor::Union{typeof(CM), typeof(ATE), typeof(AIE)},
     treatments, outcomes; 
     dataset=nothing,
     confounders=nothing, 
@@ -437,7 +439,7 @@ function factorialEstimands(
     return estimands
 end
 
-joint_levels(Ψ::StatisticalIATE) = Iterators.product(values(Ψ.treatment_values)...)
+joint_levels(Ψ::StatisticalAIE) = Iterators.product(values(Ψ.treatment_values)...)
 
 joint_levels(Ψ::StatisticalATE) =
     (Tuple(Ψ.treatment_values[T][c] for T ∈ keys(Ψ.treatment_values)) for c in (:control, :case))

src/estimands.jl

@@ -26,7 +26,7 @@ AbsentLevelError(treatment_name, val, levels) = ArgumentError(string(
 
 Checks the case/control values defining the treatment contrast are present in the dataset levels. 
 
-Note: This method is for estimands like the ATE or IATE that have case/control treatment settings represented as 
+Note: This method is for estimands like the ATE or AIE that have case/control treatment settings represented as 
 `NamedTuple`.
 """
 function check_treatment_settings(settings::NamedTuple, levels, treatment_name)

test/adjustment.jl

@@ -13,7 +13,7 @@ using TMLE
         CM(outcome=:Y₁, treatment_values=(T₁=1,)),
         ATE(outcome=:Y₁, treatment_values=(T₁=(case=1, control=0),)),
         ATE(outcome=:Y₁, treatment_values=(T₁=(case=1, control=0), T₂=(case=1, control=0))),
-        IATE(outcome=:Y₁, treatment_values=(T₁=(case=1, control=0), T₂=(case=1, control=0))),
+        AIE(outcome=:Y₁, treatment_values=(T₁=(case=1, control=0), T₂=(case=1, control=0))),
     ]
     method = BackdoorAdjustment(outcome_extra_covariates=[:C])
     statistical_estimands = [identify(method, estimand, scm) for estimand in causal_estimands]

test/composition.jl

@@ -162,25 +162,25 @@ end
         T₂ = categorical(T₂),
         Y = Y
     )
-    IATE₁ = IATE(
+    AIE₁ = AIE(
         outcome = :Y,
         treatment_values = (T₁=(case=2, control=1), T₂=(case=2, control=1)),
         treatment_confounders = (T₁ = [:W], T₂ = [:W])
     )
-    IATE₂ = IATE(
+    AIE₂ = AIE(
         outcome = :Y,
         treatment_values = (T₁=(case=3, control=1), T₂=(case=3, control=1)),
         treatment_confounders = (T₁ = [:W], T₂ = [:W])
     )
-    IATE₃ = IATE(
+    AIE₃ = AIE(
         outcome = :Y,
         treatment_values = (T₁=(case=3, control=2), T₂=(case=3, control=2)),
         treatment_confounders = (T₁ = [:W], T₂ = [:W])
     )
-    jointIATE = JointEstimand(IATE₁, IATE₂, IATE₃)
+    jointAIE = JointEstimand(AIE₁, AIE₂, AIE₃)
 
     ose = OSE(models=TMLE.default_models(G=LogisticClassifier(), Q_continuous=LinearRegressor()))
-    jointEstimate, _ = ose(jointIATE, dataset, verbosity=0)
+    jointEstimate, _ = ose(jointAIE, dataset, verbosity=0)
 
     testres = significance_test(jointEstimate)
     @test testres.x̄ ≈ estimate(jointEstimate)

test/configuration.jl

@@ -22,7 +22,7 @@ end
     yamlfilename = mktemp()[1]
     jsonfilename = mktemp()[1]
     estimands = [
-        IATE(
+        AIE(
             outcome=:Y1, 
             treatment_values= (
                 T1 = (case = 1, control = 0), 
@@ -81,7 +81,7 @@ end
     jsonfilename = mktemp()[1]
     # With a StaticSCM, some Causal estimands and an Adjustment Method
     estimands = [
-        IATE(
+        AIE(
             outcome=:Y1, 
             treatment_values= (
                 T1 = (case = 1, control = 0), 

test/counterfactual_mean_based/3points_interactions.jl

@@ -27,7 +27,7 @@ end
 
 @testset "Test 3-points interactions" begin
     dataset, Ψ₀ = dataset_scm_and_truth(;n=1000)
-    Ψ = IATE(
+    Ψ = AIE(
         outcome   = :Y,
         treatment_values = (
             T₁=(case=true, control=false), 

test/counterfactual_mean_based/clever_covariate.jl

@@ -53,7 +53,7 @@ end
 end
 
 @testset "Test clever_covariate_and_weights: 2 treatments" begin
-    Ψ = IATE(
+    Ψ = AIE(
         outcome = :Y,
         treatment_values=(
             T₁=(case=1, control=0), 
@@ -90,7 +90,7 @@ end
 
 @testset "Test compute_offset, clever_covariate_and_weights: 3 treatments" begin
     ## Third case: 3 Treatment variables
-    Ψ = IATE(
+    Ψ = AIE(
         outcome =:Y, 
         treatment_values=(
             T₁=(case="a", control="b"),