diff --git a/src/schema.jl b/src/schema.jl
index 7245793a..2f352e8f 100644
--- a/src/schema.jl
+++ b/src/schema.jl
@@ -184,9 +184,7 @@ concrete_term(t, d, hint) = t
 
 concrete_term(t::Term, xs::AbstractVector{<:Number}, ::Nothing) = concrete_term(t, xs, ContinuousTerm)
 function concrete_term(t::Term, xs::AbstractVector, ::Type{ContinuousTerm})
-    μ, σ2 = StatsBase.mean_and_var(xs)
-    min, max = extrema(xs)
-    ContinuousTerm(t.sym, promote(μ, σ2, min, max)...)
+    ContinuousTerm(t.sym)
 end
 # default contrasts: dummy coding
 concrete_term(t::Term, xs::AbstractVector, ::Nothing) = concrete_term(t, xs, CategoricalTerm)
diff --git a/src/terms.jl b/src/terms.jl
index b49a9027..f4d34cb0 100644
--- a/src/terms.jl
+++ b/src/terms.jl
@@ -201,17 +201,9 @@ Represents a continuous variable, with a name and summary statistics.
 # Fields
 
 * `sym::Symbol`: The name of the variable
-* `mean::T`: Mean
-* `var::T`: Variance
-* `min::T`: Minimum value
-* `max::T`: Maximum value
 """
-struct ContinuousTerm{T} <: AbstractTerm
+struct ContinuousTerm <: AbstractTerm
     sym::Symbol
-    mean::T
-    var::T
-    min::T
-    max::T
 end
 width(::ContinuousTerm) = 1
 
diff --git a/test/terms.jl b/test/terms.jl
index ba4cb2d0..5ddc7c3f 100644
--- a/test/terms.jl
+++ b/test/terms.jl
@@ -26,10 +26,6 @@ StatsModels.apply_schema(mt::MultiTerm, sch::StatsModels.Schema, Mod::Type) =
         t0 = concrete_term(t, [3, 2, 1])
         @test string(t0) == "aaa"
         @test mimestring(t0) == "aaa(continuous)"
-        @test t0.mean == 2.0
-        @test t0.var == var([1,2,3])
-        @test t0.min == 1.0
-        @test t0.max == 3.0
 
         t1 = concrete_term(t, [:a, :b, :c])
         @test t1.contrasts isa StatsModels.ContrastsMatrix{DummyCoding}