Merge pull request #103 from TARGENE/significance_test

Significance test

Project.toml

@@ -1,7 +1,7 @@
 name = "TMLE"
 uuid = "8afdd2fb-6e73-43df-8b62-b1650cd9c8cf"
 authors = ["Olivier Labayle"]
-version = "0.14.0"
+version = "0.14.1"
 
 [deps]
 AbstractDifferentiation = "c29ec348-61ec-40c8-8164-b8c60e9d9f3d"
@@ -19,7 +19,6 @@ MLJModels = "d491faf4-2d78-11e9-2867-c94bc002c0b7"
 MetaGraphsNext = "fa8bd995-216d-47f1-8a91-f3b68fbeb377"
 Missings = "e1d29d7a-bbdc-5cf2-9ac0-f12de2c33e28"
 PrecompileTools = "aea7be01-6a6a-4083-8856-8a6e6704d82a"
-PrettyTables = "08abe8d2-0d0c-5749-adfa-8a2ac140af0d"
 Random = "9a3f8284-a2c9-5f02-9a11-845980a1fd5c"
 SplitApplyCombine = "03a91e81-4c3e-53e1-a0a4-9c0c8f19dd66"
 Statistics = "10745b16-79ce-11e8-11f9-7d13ad32a3b2"
@@ -51,12 +50,11 @@ MLJModels = "0.15, 0.16"
 MetaGraphsNext = "0.7"
 Missings = "1.0"
 PrecompileTools = "1.1.1"
-PrettyTables = "2.2"
+SplitApplyCombine = "1.2.2"
 TableOperations = "1.2"
 Tables = "1.6"
 YAML = "0.4.9"
 Zygote = "0.6.69"
-SplitApplyCombine = "1.2.2"
 julia = "1.6, 1.7, 1"
 
 [extras]

src/TMLE.jl

@@ -14,7 +14,6 @@ using Distributions
 using Zygote
 using LogExpFunctions
 using PrecompileTools
-using PrettyTables
 using Random
 import AbstractDifferentiation as AD
 using Graphs
@@ -32,7 +31,8 @@ export AVAILABLE_ESTIMANDS
 export factorialATE, factorialIATE
 export TMLEE, OSE, NAIVE
 export ComposedEstimand
-export var, estimate, OneSampleTTest, OneSampleZTest, OneSampleHotellingT2Test,pvalue, confint, emptyIC
+export var, estimate, pvalue, confint, emptyIC
+export significance_test, OneSampleTTest, OneSampleZTest, OneSampleHotellingT2Test
 export compose
 export TreatmentTransformer, with_encoder, encoder
 export BackdoorAdjustment, identify

src/counterfactual_mean_based/estimates.jl

@@ -65,13 +65,6 @@ end
 
 emptyIC(estimate; pval_threshold=nothing) = emptyIC(estimate, pval_threshold)
 
-
-function Base.show(io::IO, ::MIME"text/plain", est::EICEstimate)
-    testresult = OneSampleTTest(est)
-    data = [estimate(est) confint(testresult) pvalue(testresult);]
-    pretty_table(io, data;header=["Estimate", "95% Confidence Interval", "P-value"])
-end
-
 """
     Distributions.estimate(r::EICEstimate)
 
@@ -104,3 +97,12 @@ Performs a T test on the EICEstimate.
 HypothesisTests.OneSampleTTest(est::EICEstimate, Ψ₀=0) = 
     OneSampleTTest(est.estimate, est.std, est.n, Ψ₀)
 
+"""
+    significance_test(estimate::EICEstimate, Ψ₀=0)
+
+Performs a TTest
+"""
+significance_test(estimate::EICEstimate, Ψ₀=0) = OneSampleTTest(estimate, Ψ₀)
+
+Base.show(io::IO, mime::MIME"text/plain", est::Union{EICEstimate, ComposedEstimand}) =
+    show(io, mime, significance_test(est))

src/estimates.jl

@@ -109,18 +109,6 @@ to_matrix(x) = reduce(hcat, x)
 ComposedEstimate(;estimand, estimates, estimate, cov, n) =
     ComposedEstimate(estimand, Tuple(estimates), collect(estimate), to_matrix(cov), n)
 
-
-function Base.show(io::IO, ::MIME"text/plain", est::ComposedEstimate)
-    if length(est.cov) !== 1
-        println(io, string("Estimate: ", estimate(est), "\nVariance: \n", var(est)))
-    else
-        testresult = OneSampleTTest(est)
-        data = [estimate(est) confint(testresult) pvalue(testresult);]
-        headers = ["Estimate", "95% Confidence Interval", "P-value"]
-        pretty_table(io, data;header=headers)
-    end
-end
-
 """
     Distributions.estimate(r::ComposedEstimate)
 
@@ -171,6 +159,20 @@ function HypothesisTests.OneSampleZTest(estimate::ComposedEstimate, Ψ₀=0)
     return OneSampleZTest(estimate.estimate[1], sqrt(estimate.cov[1]), estimate.n, Ψ₀)
 end
 
+"""
+    significance_test(estimate::ComposedEstimate, Ψ₀=zeros(size(estimate.estimate, 1)))
+
+Performs a TTest if the estimate is one dimensional and a HotellingT2Test otherwise.
+"""
+function significance_test(estimate::ComposedEstimate, Ψ₀=zeros(size(estimate.estimate, 1)))
+    if length(estimate.estimate) == 1
+        Ψ₀ = Ψ₀ isa AbstractArray ? first(Ψ₀) : Ψ₀
+        return OneSampleTTest(estimate, Ψ₀)
+    else
+        return OneSampleHotellingT2Test(estimate, Ψ₀)
+    end
+end
+
 function emptyIC(estimate::ComposedEstimate, pval_threshold)
     emptied_estimates = Tuple(emptyIC(e, pval_threshold) for e in estimate.estimates)
     ComposedEstimate(estimate.estimand, emptied_estimates, estimate.estimate, estimate.cov, estimate.n)

test/composition.jl

@@ -200,7 +200,7 @@ end
     ose = OSE(models=TMLE.default_models(G=LogisticClassifier(), Q_continuous=LinearRegressor()))
     jointEstimate, _ = ose(jointIATE, dataset, verbosity=0)
 
-    testres = OneSampleHotellingT2Test(jointEstimate)
+    testres = significance_test(jointEstimate)
     @test testres.x̄ ≈ jointEstimate.estimate
     @test pvalue(testres) < 1e-10
 
@@ -213,7 +213,7 @@ end
     maybe_emptied_estimate = TMLE.emptyIC(jointEstimate, pval_threshold=pval_threshold)
     n_empty = 0
     for i in 1:3
-        pval = pvalue(OneSampleTTest(jointEstimate.estimates[i]))
+        pval = pvalue(significance_test(jointEstimate.estimates[i]))
         maybe_emptied_IC = maybe_emptied_estimate.estimates[i].IC
         if pval > pval_threshold
             @test maybe_emptied_IC == []

test/counterfactual_mean_based/non_regression_test.jl

@@ -12,7 +12,7 @@ using YAML
 
 function regression_tests(tmle_result)
     @test estimate(tmle_result) ≈ -0.185533 atol = 1e-6
-    l, u = confint(OneSampleTTest(tmle_result))
+    l, u = confint(significance_test(tmle_result))
     @test l ≈ -0.279246 atol = 1e-6
     @test u ≈ -0.091821 atol = 1e-6
     @test OneSampleZTest(tmle_result) isa OneSampleZTest

test/helper_fns.jl

@@ -39,10 +39,7 @@ at the given confidence level: here 0.05
 """
 function test_coverage(result::TMLE.EICEstimate, Ψ₀)
     # TMLE
-    lb, ub = confint(OneSampleTTest(result))
-    @test lb ≤ Ψ₀ ≤ ub
-    # OneStep
-    lb, ub = confint(OneSampleZTest(result))
+    lb, ub = confint(significance_test(result))
     @test lb ≤ Ψ₀ ≤ ub
 end