diff --git a/src/univariate/continuous/gamma.jl b/src/univariate/continuous/gamma.jl
index 16d96088c..5b60b6811 100644
--- a/src/univariate/continuous/gamma.jl
+++ b/src/univariate/continuous/gamma.jl
@@ -1,13 +1,14 @@
 immutable Gamma <: ContinuousUnivariateDistribution
     α::Float64
-    β::Float64
+    θ::Float64
 
-    function Gamma(α::Real, β::Real)
-        α > zero(α) && β > zero(β) || error("Gamma: both shape and scale must be positive")
-        @compat new(Float64(α), Float64(β))
+    function Gamma(α::Real, θ::Real)
+        α > zero(α) && θ > zero(θ) ||
+            throw(ArgumentError("Gamma: both α and Θ must be positive"))
+        @compat new(Float64(α), Float64(θ))
     end
 
-    Gamma(α::Real) = Gamma(α, 1.0)
+    Gamma(α::Real) = @compat Gamma(Float64(α), 1.0)
     Gamma() = new(1.0, 1.0)
 end
 
@@ -17,45 +18,45 @@ end
 #### Parameters
 
 shape(d::Gamma) = d.α
-scale(d::Gamma) = d.β
-rate(d::Gamma) = 1.0 / d.β
+scale(d::Gamma) = d.θ
+rate(d::Gamma) = 1.0 / d.θ
 
-params(d::Gamma) = (d.α, d.β)
+params(d::Gamma) = (d.α, d.θ)
 
 
 #### Statistics
 
-mean(d::Gamma) = d.α * d.β
+mean(d::Gamma) = d.α * d.θ
 
-var(d::Gamma) = d.α * d.β^2
+var(d::Gamma) = d.α * d.θ^2
 
 skewness(d::Gamma) = 2.0 / sqrt(d.α)
 
 kurtosis(d::Gamma) = 6.0 / d.α
 
 function mode(d::Gamma)
-    (α, β) = params(d)
-    α >= 1.0 ? β * (α - 1.0) : error("Gamma has no mode when shape < 1.0")
+    (α, θ) = params(d)
+    α >= 1.0 ? θ * (α - 1.0) : error("Gamma has no mode when shape < 1.0")
 end
 
 function entropy(d::Gamma)
-    (α, β) = params(d)
-    α + lgamma(α) + (1.0 - α) * digamma(α) + log(β)
+    (α, θ) = params(d)
+    α + lgamma(α) + (1.0 - α) * digamma(α) + log(θ)
 end
 
-mgf(d::Gamma, t::Real) = (1.0 - t * d.β)^(-d.α)
+mgf(d::Gamma, t::Real) = (1.0 - t * d.θ)^(-d.α)
 
-cf(d::Gamma, t::Real) = (1.0 - im * t * d.β)^(-d.α)
+cf(d::Gamma, t::Real) = (1.0 - im * t * d.θ)^(-d.α)
 
 
 #### Evaluation & Sampling
 
-@_delegate_statsfuns Gamma gamma α β
+@_delegate_statsfuns Gamma gamma α θ
 
 gradlogpdf(d::Gamma, x::Float64) =
-    insupport(Gamma, x) ? (d.α - 1.0) / x - 1.0 / d.β : 0.0
+    insupport(Gamma, x) ? (d.α - 1.0) / x - 1.0 / d.θ : 0.0
 
-rand(d::Gamma) = StatsFuns.Rmath.gammarand(d.α, d.β)
+rand(d::Gamma) = StatsFuns.Rmath.gammarand(d.α, d.θ)
 
 
 #### Fit model
diff --git a/src/univariate/continuous/gumbel.jl b/src/univariate/continuous/gumbel.jl
index 55de73416..2e6128eab 100644
--- a/src/univariate/continuous/gumbel.jl
+++ b/src/univariate/continuous/gumbel.jl
@@ -1,13 +1,14 @@
 immutable Gumbel <: ContinuousUnivariateDistribution
-    μ::Float64   # location
-    β::Float64 # scale
+    μ::Float64  # location
+    θ::Float64  # scale
 
-    function Gumbel(μ::Real, β::Real)
-        β > zero(β) || error("The scale of Gumbel must be positive")
-        @compat new(Float64(μ), Float64(β))
+    function Gumbel(μ::Real, θ::Real)
+        θ > zero(θ) ||
+            throw(ArgumentError("Gumbel: Θ must be positive."))
+        @compat new(Float64(μ), Float64(θ))
     end
 
-    Gumbel(μ::Real) = Gumbel(μ, 1.0)
+    Gumbel(μ::Real) = @compat Gumbel(Float64(μ), 1.0)
     Gumbel() = new(0.0, 1.0)
 end
 
@@ -19,52 +20,50 @@ const DoubleExponential = Gumbel
 #### Parameters
 
 location(d::Gumbel) = d.μ
-scale(d::Gumbel) = d.β
-params(d::Gumbel) = (d.μ, d.β)
+scale(d::Gumbel) = d.θ
+params(d::Gumbel) = (d.μ, d.θ)
 
 
 #### Statistics
 
-mean(d::Gumbel) = d.μ + d.β * 0.57721566490153286
+mean(d::Gumbel) = d.μ + d.θ * 0.57721566490153286
 
-median(d::Gumbel) = d.μ + d.β * 0.366512920581664327
+median(d::Gumbel) = d.μ + d.θ * 0.366512920581664327
 
 mode(d::Gumbel) = d.μ
 
-var(d::Gumbel) = 1.6449340668482264 * d.β^2
+var(d::Gumbel) = 1.6449340668482264 * d.θ^2
 
 skewness(d::Gumbel) = 1.13954709940464866
 
 kurtosis(d::Gumbel) = 2.4
 
-entropy(d::Gumbel) = 1.57721566490153286 + log(d.β)
+entropy(d::Gumbel) = 1.57721566490153286 + log(d.θ)
 
 
 #### Evaluation
 
-zval(d::Gumbel, x::Float64) = (x - d.μ) / d.β
-xval(d::Gumbel, z::Float64) = x * d.β + d.μ
+zval(d::Gumbel, x::Float64) = (x - d.μ) / d.θ
+xval(d::Gumbel, z::Float64) = x * d.θ + d.μ
 
 function pdf(d::Gumbel, x::Float64)
     z = zval(d, x)
-    exp(-z - exp(-z)) / d.β
+    exp(-z - exp(-z)) / d.θ
 end
 
 function logpdf(d::Gumbel, x::Float64)
     z = zval(d, x)
-    - (z + exp(-z) + log(d.β))
+    - (z + exp(-z) + log(d.θ))
 end
 
 cdf(d::Gumbel, x::Float64) = exp(-exp(-zval(d, x)))
 logcdf(d::Gumbel, x::Float64) = -exp(-zval(d, x))
 
-quantile(d::Gumbel, p::Float64) = d.μ - d.β * log(-log(p))
+quantile(d::Gumbel, p::Float64) = d.μ - d.θ * log(-log(p))
 
-gradlogpdf(d::Gumbel, x::Float64) = - (1.0 + exp((d.μ - x) / d.β)) / d.β
+gradlogpdf(d::Gumbel, x::Float64) = - (1.0 + exp((d.μ - x) / d.θ)) / d.θ
 
 
 #### Sampling
 
 rand(d::Gumbel) = quantile(d, rand())
-
-
diff --git a/src/univariate/continuous/inversegamma.jl b/src/univariate/continuous/inversegamma.jl
index 4aa6b98ee..5403b660f 100644
--- a/src/univariate/continuous/inversegamma.jl
+++ b/src/univariate/continuous/inversegamma.jl
@@ -1,13 +1,14 @@
 immutable InverseGamma <: ContinuousUnivariateDistribution
     invd::Gamma
-    β::Float64
+    θ::Float64
 
-    function InverseGamma(α::Real, β::Real)
-        (α > zero(α) && β > zero(β)) || error("Both shape and scale must be positive.")
-        @compat new(Gamma(α, 1.0 / β), Float64(β))
+    function InverseGamma(α::Real, θ::Real)
+        (α > zero(α) && θ > zero(θ)) ||
+            throw(ArgumentError("InverseGamma: both α and θ must be positive."))
+        @compat new(Gamma(α, 1.0 / θ), Float64(θ))
     end
 
-    InverseGamma(α::Real) = InverseGamma(α, 1.0)
+    InverseGamma(α::Real) = @compat InverseGamma(Float64(α), 1.0)
     InverseGamma() = InverseGamma(1.0, 1.0)
 end
 
@@ -17,7 +18,7 @@ end
 #### Parameters
 
 shape(d::InverseGamma) = shape(d.invd)
-scale(d::InverseGamma) = d.β
+scale(d::InverseGamma) = d.θ
 rate(d::InverseGamma) = scale(d.invd)
 
 params(d::InverseGamma) = (shape(d), scale(d))
@@ -25,13 +26,13 @@ params(d::InverseGamma) = (shape(d), scale(d))
 
 #### Parameters
 
-mean(d::InverseGamma) = ((α, β) = params(d); α  > 1.0 ? β / (α - 1.0) : Inf)
+mean(d::InverseGamma) = ((α, θ) = params(d); α  > 1.0 ? θ / (α - 1.0) : Inf)
 
 mode(d::InverseGamma) = scale(d) / (shape(d) + 1.0)
 
 function var(d::InverseGamma)
-    (α, β) = params(d)
-    α > 2.0 ? β^2 / ((α - 1.0)^2 * (α - 2.0)) : Inf
+    (α, θ) = params(d)
+    α > 2.0 ? θ^2 / ((α - 1.0)^2 * (α - 2.0)) : Inf
 end
 
 function skewness(d::InverseGamma)
@@ -45,8 +46,8 @@ function kurtosis(d::InverseGamma)
 end
 
 function entropy(d::InverseGamma)
-    (α, β) = params(d)
-    α + lgamma(α) - (1.0 + α) * digamma(α) + log(β)
+    (α, θ) = params(d)
+    α + lgamma(α) - (1.0 + α) * digamma(α) + log(θ)
 end
 
 
@@ -55,8 +56,8 @@ end
 pdf(d::InverseGamma, x::Float64) = exp(logpdf(d, x))
 
 function logpdf(d::InverseGamma, x::Float64)
-    (α, β) = params(d)
-    α * log(β) - lgamma(α) - (α + 1.0) * log(x) - β / x
+    (α, θ) = params(d)
+    α * log(θ) - lgamma(α) - (α + 1.0) * log(x) - θ / x
 end
 
 cdf(d::InverseGamma, x::Float64) = ccdf(d.invd, 1.0 / x)
@@ -92,4 +93,3 @@ function _rand!(d::InverseGamma, A::AbstractArray)
     end
     A
 end
-
diff --git a/src/univariate/continuous/inversegaussian.jl b/src/univariate/continuous/inversegaussian.jl
index ad51b0998..27be1c222 100644
--- a/src/univariate/continuous/inversegaussian.jl
+++ b/src/univariate/continuous/inversegaussian.jl
@@ -3,11 +3,12 @@ immutable InverseGaussian <: ContinuousUnivariateDistribution
     λ::Float64
 
     function InverseGaussian(μ::Real, λ::Real)
-        (μ > zero(μ) && λ > zero(λ)) || error("InverseGaussian's μ and λ must be positive")
+        (μ > zero(μ) && λ > zero(λ)) ||
+            throw(ArgumentError("InverseGaussian: μ and λ must be positive."))
         @compat new(Float64(μ), Float64(λ))
     end
 
-    InverseGaussian(μ::Real) = InverseGaussian(μ, 1.0)
+    InverseGaussian(μ::Real) = @compat InverseGaussian(Float64(μ), 1.0)
     InverseGaussian() = new(1.0, 1.0)
 end