Release 0.2.5

Project.toml

@@ -1,7 +1,7 @@
 name = "AnovaFixedEffectModels"
 uuid = "e4965305-65d6-464d-9c03-ae3e5cffadab"
 authors = ["yufongpeng <54415349+yufongpeng@users.noreply.github.com> and contributors"]
-version = "0.2.4"
+version = "0.2.5"
 
 [deps]
 AnovaBase = "946dddda-6a23-4b48-8e70-8e60d9b8d680"
@@ -16,11 +16,11 @@ StatsModels = "3eaba693-59b7-5ba5-a881-562e759f1c8d"
 Tables = "bd369af6-aec1-5ad0-b16a-f7cc5008161c"
 
 [compat]
-AnovaBase = "0.7"
+AnovaBase = "0.8"
 Distributions = "0.23, 0.24, 0.25"
 FixedEffectModels = "1.3, 1.4, 1.5, 1.6, 1.7, 1.8"
 Reexport = "0.2, 1"
 StatsBase = "0.33, 0.34"
 StatsModels = "0.7"
 Tables = "1.7"
-julia = "1.6, 1.7, 1.8, 1.9"
+julia = "1.6, 1.7, 1.8, 1.9, 1.10"

src/AnovaFixedEffectModels.jl

@@ -7,10 +7,10 @@ import StatsBase: fit!, fit
 import StatsModels: TableRegressionModel, vectorize, width, apply_schema, 
                     ModelFrame, ModelMatrix, columntable, asgn
 
-using AnovaBase: select_super_interaction, extract_contrasts, canonicalgoodnessoffit, subformula, dof_asgn, lrt_nested, ftest_nested, _diff, _diffn, has_intercept
+using AnovaBase: select_super_interaction, extract_contrasts, canonicalgoodnessoffit, subformula, dof_asgn, lrt_nested, ftest_nested, _diff, _diffn, testname
 import AnovaBase: anova, nestedmodels, anovatable, prednames, predictors, formula
 using Tables: columntable
-
+import Base: show
 export anova_lfe, lfe
 
 include("anova.jl")

src/anova.jl

@@ -66,10 +66,10 @@ function anova(::Type{FTest},
         end
     elseif aovm.type == 2
         fstat = ntuple(last(fullasgn) - offset) do fix
-            select1 = sort!(collect(select_super_interaction(fullpred, fix + offset)))
-            select2 = setdiff(select1, fix + offset)
-            select1 = findall(in(select1), fullasgn)
-            select2 = findall(in(select2), fullasgn)
+            s1 = sort!(collect(select_super_interaction(fullpred, fix + offset)))
+            s2 = setdiff(s1, fix + offset)
+            select1 = findall(in(s1), fullasgn)
+            select2 = findall(in(s2), fullasgn)
             (β[select1]' * (varβ[select1, select1] \ β[select1]) - β[select2]' * (varβ[select2, select2] \ β[select2])) / df[fix]
         end
     else
@@ -83,7 +83,7 @@ function anova(::Type{FTest},
     σ² = rss(aovm.model) / dfr
     devs = @. fstat * σ² * df
     pvalue = @. ccdf(FDist(df, dfr), abs(fstat))
-    AnovaResult{FTest}(aovm, df, devs, fstat, pvalue, NamedTuple())
+    AnovaResult(aovm, FTest, df, devs, fstat, pvalue, NamedTuple())
 end
 
 # =================================================================================================================
@@ -102,7 +102,7 @@ function anova(::Type{FTest},
     dev = rss.(models)
     # check comparable and nested
     check && @warn "Could not check whether models are nested: results may not be meaningful"
-    ftest_nested(NestedModels{M}(models), df, dfr, dev, last(dev) / last(dfr))
+    ftest_nested(NestedModels(models), df, dfr, dev, last(dev) / last(dfr))
 end
 
 function anova(::Type{FTest}, aovm::NestedModels{M}) where {M <: FixedEffectModel}

src/fit.jl

@@ -1,6 +1,5 @@
 # ==========================================================================================================
 # Backend funcion
-formula(model::T) where {T <: FixedEffectModel} = model.formula
 predictors(model::T) where {T <: FixedEffectModel} = model.formula_schema.rhs.terms
 
 # Variable dispersion
@@ -19,10 +18,10 @@ Generate nested models from modeltype, formula and data. The null model will be
 function nestedmodels(modeltype::Type{FixedEffectModel}, f::FormulaTerm, data; null = true, kwargs...)
     fullm = lfe(f, data; kwargs...)
     predterms = predictors(fullm)
-    has_intercept(predterms) || (length(predterms) > 1 ? (predterms = predterms[2:end]) : throw(ArgumentError("Empty model is given!")))
+    hasintercept(predterms) || (length(predterms) > 1 ? (predterms = predterms[2:end]) : throw(ArgumentError("Empty model is given!")))
     feterms = filter(isfe, f.rhs)
     subm = ntuple(length(predterms) - 1) do i
         lfe(FormulaTerm(f.lhs, (predterms[1:i]..., feterms...)), data; kwargs...)
     end
-    NestedModels{FixedEffectModel}(null ? (lfe(FormulaTerm(f.lhs, (ConstantTerm(0), feterms...)), data; kwargs...), subm..., fullm) : (subm..., fullm))
+    NestedModels(null ? (lfe(FormulaTerm(f.lhs, (ConstantTerm(0), feterms...)), data; kwargs...), subm..., fullm) : (subm..., fullm))
 end

src/io.jl

@@ -38,4 +38,59 @@ function anovatable(aov::AnovaResult{<: NestedModels{<: FixedEffectModel, N}, FT
                 ],
                 ["DOF", "ΔDOF", "Res.DOF", "R²", "ΔR²", "R²_within", "ΔR²_within", "Res.SS", "Exp.SS", "F value", "Pr(>|F|)"],
                 rownames, 11, 10)
-end 
+end 
+
+function show(io::IO, anovamodel::FullModel{<: T}) where {T <: FixedEffectModel}
+    println(io, "FullModel for type $(anovamodel.type) test")
+    println(io)
+    println(io, "Predictors:")
+    println(io, join(prednames(anovamodel), ", "))
+    println(io)
+    println(io, "Formula:")
+    println(io, anovamodel.model.formula)
+    println(io)
+    println(io, "Coefficients:")
+    show(io, coeftable(anovamodel.model))
+end
+
+function show(io::IO, anovamodel::NestedModels{M, N}) where {M <: FixedEffectModel, N}
+    println(io, "NestedModels with $N models")
+    println(io)
+    println(io, "Formulas:")
+    for(id, m) in enumerate(anovamodel.model)
+        println(io, "Model $id: ", m.formula)
+    end
+    println(io)
+    println(io, "Coefficients:")
+    show(io, coeftable(first(anovamodel.model)))
+    println(io)
+    N > 2 && print(io, " .\n" ^ 3)
+    show(io, coeftable(last(anovamodel.model)))
+end
+
+# Show function that delegates to anovatable
+function show(io::IO, aov::AnovaResult{<: FullModel{<: FixedEffectModel}, T}) where {T <: GoodnessOfFit}
+    at = anovatable(aov)
+    println(io, "Analysis of Variance")
+    println(io)
+    println(io, "Type $(anova_type(aov)) test / $(testname(T))")
+    println(io)
+    println(io, aov.anovamodel.model.formula)
+    println(io)
+    println(io, "Table:")
+    show(io, at)
+end
+
+function show(io::IO, aov::AnovaResult{<: MultiAovModels{<: FixedEffectModel}, T}) where {T <: GoodnessOfFit}
+    at = anovatable(aov)
+    println(io,"Analysis of Variance")
+    println(io)
+    println(io, "Type $(anova_type(aov)) test / $(testname(T))")
+    println(io)
+    for(id, m) in enumerate(aov.anovamodel.model)
+        println(io, "Model $id: ", m.formula)
+    end
+    println(io)
+    println(io, "Table:")
+    show(io, at)
+end

test/runtests.jl

@@ -52,12 +52,14 @@ isapprox(x::NTuple{N, Float64}, y::NTuple{N, Float64}, atol::NTuple{N, Float64}
             global aovl1 = AnovaGLM.anova(lm1)
             global aovf2 = AFE.anova(fem2, type = 3)
             global aovl2 = AnovaGLM.anova(lm2, type = 3)
-            global aovfs = AFE.anova(NestedModels{FixedEffectModel}(fem0, fem1))
+            global aovfs = AFE.anova(NestedModels(fem0, fem1))
             global aovfs2 = AFE.anova(fem0, fem1)
             global aovls = AnovaGLM.anova(lm0, lm1)
             @test !(@test_error test_show(aovf1))
+            @test !(@test_error test_show(aovf1.anovamodel))
             @test !(@test_error test_show(aovf2))
             @test !(@test_error test_show(aovfs))
+            @test !(@test_error test_show(aovfs.anovamodel))
             @test nobs(aovf1) == nobs(aovl1)
             @test last(dof(aovf1)) == dof(aovl1)[end - 1]
             @test isapprox(first(deviance(aovf2)), first(deviance(aovl2)))
@@ -68,7 +70,6 @@ isapprox(x::NTuple{N, Float64}, y::NTuple{N, Float64}, atol::NTuple{N, Float64}
             df = DataFrame(y = randn(1000), x = rand(1:5, 1000), z = rand(["1", "2"], 1000), t = 1:1000)
             fems1 = nestedmodels(FixedEffectModel, @formula(y ~ t + fe(z) + fe(x)), df)
             fems2 = nestedmodels(FixedEffectModel, @formula(y ~ z + t & fe(x)), df)
-            @test formula(fems1.model[1]).rhs == @formula(y ~ 0 + fe(z) + fe(x)).rhs
             @test AFE.predictors(fems2.model[2])[1] == InterceptTerm{true}()
         end
     end