all specifying type argument n GLMMi predict on original data

Author: palday

src/predict.jl

@@ -86,6 +86,13 @@ function StatsAPI.predict(
     return type == :linpred ? y : broadcast!(Base.Fix1(linkinv, Link(m)), y, y)
 end
 
+function StatsAPI.predict(m::GeneralizedLinearMixedModel; type=:response)
+    type in (:linpred, :response) || throw(ArgumentError("Invalid value for type: $(type)"))
+    y = fitted(m)
+    type == :response ? y : broadcast!(Base.Fix1(linkfun, Link(m)), y, y)
+    return y
+end
+
 # β is separated out here because m.β != m.LMM.β depending on how β is estimated for GLMM
 # also β should already be pivoted but NOT truncated in the rank deficient case
 function _predict(m::MixedModel{T}, newdata, β; new_re_levels) where {T}

test/predict.jl

@@ -197,6 +197,8 @@ end
             # we can skip a lot of testing if the broad strokes work because
             # internally this is punted off to the same machinery as LMM
             @test predict(gm0) ≈ fitted(gm0)
+            @test predict(gm0; type=:linpred) ≈ predict(gm0, contra; type=:linpred)
+            @test predict(gm0; type=:response) ≈ predict(gm0, contra; type=:response)
             # XXX these tolerances aren't great but are required for fast=false fits
             @test predict(gm0, contra; type=:linpred) ≈ gm0.resp.eta rtol = 0.1
             @test predict(gm0, contra; type=:response) ≈ gm0.resp.mu rtol = 0.01