support intervals

Project.toml

@@ -7,6 +7,7 @@ version = "0.25.104"
 ChainRulesCore = "d360d2e6-b24c-11e9-a2a3-2a2ae2dbcce4"
 DensityInterface = "b429d917-457f-4dbc-8f4c-0cc954292b1d"
 FillArrays = "1a297f60-69ca-5386-bcde-b61e274b549b"
+IntervalSets = "8197267c-284f-5f27-9208-e0e47529a953"
 LinearAlgebra = "37e2e46d-f89d-539d-b4ee-838fcccc9c8e"
 PDMats = "90014a1f-27ba-587c-ab20-58faa44d9150"
 Printf = "de0858da-6303-5e67-8744-51eddeeeb8d7"
@@ -39,6 +40,7 @@ Distributed = "<0.0.1, 1"
 FillArrays = "0.9, 0.10, 0.11, 0.12, 0.13, 1"
 FiniteDifferences = "0.12"
 ForwardDiff = "0.10"
+IntervalSets = "0.7"
 JSON = "0.21"
 LinearAlgebra = "<0.0.1, 1"
 OffsetArrays = "1"

src/Distributions.jl

@@ -27,6 +27,10 @@ import PDMats: dim, PDMat, invquad
 using SpecialFunctions
 using Base.MathConstants: eulergamma
 
+using IntervalSets
+
+const RealInterval = Interval
+
 export
     # re-export Statistics
     mean, median, quantile, std, var, cov, cor,
@@ -328,6 +332,12 @@ end
 
 include("deprecates.jl")
 
+
+Uniform(i::Interval) = Uniform(leftendpoint(i), rightendpoint(i))
+LogUniform(i::Interval) = LogUniform(leftendpoint(i), rightendpoint(i))
+truncated(d0, i::Interval) = truncated(d, leftendpoint(i), rightendpoint(i))
+censored(d0, i::Interval) = censored(d, leftendpoint(i), rightendpoint(i))
+
 """
 A Julia package for probability distributions and associated functions.
 

src/univariate/locationscale.jl

@@ -78,11 +78,11 @@ minimum(d::AffineDistribution) =
 maximum(d::AffineDistribution) =
     d.σ > 0 ? d.μ + d.σ * maximum(d.ρ) : d.μ + d.σ * minimum(d.ρ)
 support(d::AffineDistribution) = affinedistribution_support(d.μ, d.σ, support(d.ρ))
-function affinedistribution_support(μ::Real, σ::Real, support::RealInterval)
+function affinedistribution_support(μ::Real, σ::Real, support::Interval)
     if σ > 0
-        return RealInterval(μ + σ * support.lb, μ + σ * support.ub)
+        return Interval(μ + σ * minimum(support), μ + σ * maximum(support))
     else
-        return RealInterval(μ + σ * support.ub, μ + σ * support.lb)
+        return Interval(μ + σ * maximum(support), μ + σ * minimum(support))
     end
 end
 affinedistribution_support(μ::Real, σ::Real, support) = σ > 0 ? μ .+ σ .* support : μ .+ σ .* reverse(support)

src/univariates.jl

@@ -1,17 +1,5 @@
 #### Domain && Support
 
-struct RealInterval{T<:Real}
-    lb::T
-    ub::T
-end
-
-RealInterval(lb::Real, ub::Real) = RealInterval(promote(lb, ub)...)
-
-Base.minimum(r::RealInterval) = r.lb
-Base.maximum(r::RealInterval) = r.ub
-Base.extrema(r::RealInterval) = (r.lb, r.ub)
-Base.in(x::Real, r::RealInterval) = r.lb <= x <= r.ub
-
 isbounded(d::Union{D,Type{D}}) where {D<:UnivariateDistribution} = isupperbounded(d) && islowerbounded(d)
 
 islowerbounded(d::Union{D,Type{D}}) where {D<:UnivariateDistribution} = minimum(d) > -Inf
@@ -20,8 +8,6 @@ isupperbounded(d::Union{D,Type{D}}) where {D<:UnivariateDistribution} = maximum(
 hasfinitesupport(d::Union{D,Type{D}}) where {D<:DiscreteUnivariateDistribution} = isbounded(d)
 hasfinitesupport(d::Union{D,Type{D}}) where {D<:ContinuousUnivariateDistribution} = false
 
-Base.:(==)(r1::RealInterval, r2::RealInterval) = r1.lb == r2.lb && r1.ub == r2.ub
-
 """
     params(d::UnivariateDistribution)
 
@@ -108,7 +94,7 @@ insupport(d::Union{D,Type{D}}, X::AbstractArray) where {D<:UnivariateDistributio
 insupport(d::Union{D,Type{D}},x::Real) where {D<:ContinuousUnivariateDistribution} = minimum(d) <= x <= maximum(d)
 insupport(d::Union{D,Type{D}},x::Real) where {D<:DiscreteUnivariateDistribution} = isinteger(x) && minimum(d) <= x <= maximum(d)
 
-support(d::Union{D,Type{D}}) where {D<:ContinuousUnivariateDistribution} = RealInterval(minimum(d), maximum(d))
+support(d::Union{D,Type{D}}) where {D<:ContinuousUnivariateDistribution} = Interval(minimum(d), maximum(d))
 support(d::Union{D,Type{D}}) where {D<:DiscreteUnivariateDistribution} = round(Int, minimum(d)):round(Int, maximum(d))
 
 # Type used for dispatch on finite support

test/pdfnorm.jl

@@ -5,7 +5,7 @@ using QuadGK
 # norms of the distributions.  These methods aren't very robust because can't
 # deal with divergent norms, or discrete distributions with infinite support.
 numeric_norm(d::ContinuousUnivariateDistribution) =
-    quadgk(x -> pdf(d, x) ^ 2, support(d).lb, support(d).ub)[1]
+    quadgk(x -> pdf(d, x) ^ 2, extrema(support(d))...)[1]
 
 function numeric_norm(d::DiscreteUnivariateDistribution)
     # When the distribution has infinite support, sum up to an arbitrary large