fix and test call to fit_median_quantile

src/fitstats.jl

@@ -203,6 +203,7 @@ macro qs_cf95(q0_025, q0_975)
 end
 
 # The non-macro versions return percentile whose type matches that of the argument
+qp(q,p) = QuantilePoint(q, p)
 qp_ll(q0_025::T) where {T} = QuantilePoint(q0_025, T(0.025))
 qp_l(q0_05::T) where {T} = QuantilePoint(q0_05, T(0.05))
 qp_m(median::T) where {T} = QuantilePoint(median, T(0.5))

src/univariate/continuous/logitnormal.jl

@@ -36,12 +36,12 @@ function fit_mode_quantile(::Type{LogitNormal},
     fit_mode_quantile(LogitNormal{T}, mode, qp)
 end
 function fit_mode_quantile(::Type{LogitNormal{T}}, mode::Real, qp::QuantilePoint) where {T}
-    matchModeUpper(T(mode), qp.q, Val(40); perc = qp.p)
+    matchModeUpper(T(mode), qp, Val(40))
 end
 
-function matchModeUpper(mode::T, upper, ::Val{nTry}
-    ; perc::Real = 0.99) where {nTry, T <: Real}
-    mode == 0.5 && return matchMedianUpper(LogitNormal, 0.5, upper; perc = perc)
+function matchModeUpper(mode::T, qp::QuantilePoint, ::Val{nTry}) where {nTry, T <: Real}
+    # for symmetric - same as fitting median
+    mode == 0.5 && return fit_median_quantile(LogitNormal, mode, qp)
     # for given mu we can compute sigma by mode and upper quantile
     # hence univariate search for mu
     # we now that mu is in (\code{logit(mode)},0) for \code{mode < 0.5} and in 
@@ -50,6 +50,8 @@ function matchModeUpper(mode::T, upper, ::Val{nTry}
     # hence, first get near the global minimum by a evaluating the cost at a 
     # grid that is spaced narrower at the edge
     #
+    upper = qp.q
+    perc = qp.p
     logitMode = logit(mode)
     logitUpper = logit(upper)
     upperMu = abs(logitMode) - eps()

test/Project.toml

@@ -1,5 +1,6 @@
 [deps]
 Distributions = "31c24e10-a181-5473-b8eb-7969acd0382f"
+JET = "c3a54625-cd67-489e-a8e7-0a5a0ff4e31b"
 LoggingExtras = "e6f89c97-d47a-5376-807f-9c37f3926c36"
 Optim = "429524aa-4258-5aef-a3af-852621145aeb"
 QuadGK = "1fd47b50-473d-5c70-9696-f719f8f3bcdc"

test/runtests.jl

@@ -3,6 +3,12 @@ using Test
 using Random: Random
 using LoggingExtras
 
+#using Aqua; Aqua.test_all(DistributionFits) # ambiguities from other packages
+#using JET; JET.report_package(DistributionFits) # 
+#invalid possible error due to quantile may accept/return an Array (we pass a scalar)
+#only report problems in this module:
+#using JET; JET.report_package(DistributionFits; target_modules=(@__MODULE__,)) # 
+
 @testset "optimize error" begin
     @test_throws Exception DistributionFits.optimize(x -> x * x, -1, 1)
 end
@@ -37,3 +43,11 @@ include("univariate/test_univariate.jl")
 println("Potentially stale exports: ")
 display(Test.detect_ambiguities(DistributionFits))
 println()
+
+using JET: JET
+@testset "JET" begin
+    @static if VERSION ≥ v"1.9.2"
+        JET.test_package(DistributionFits; target_modules=(@__MODULE__,)) # 
+    end    
+end;
+

test/univariate/continuous/lognormal.jl

@@ -37,3 +37,15 @@ end;
     #
     # plot(d); plot!(dfit32)
 end;
+
+@testset "fit_mode_quantile_0.5" begin
+    # special for LogitNormal mode
+    d = fit_mode_quantile(LogNormal, 0.5, @qp(0.9, 0.95))
+    @test mode(d) ≈ 0.5
+    @test quantile(d, 0.95) ≈ 0.9
+    #
+    dfit32 = fit_mode_quantile(LogNormal, 0.5f0, @qp(0.9, 0.95))
+    @test partype(dfit32) === Float32
+    @test mode(d) ≈ 0.5
+    @test quantile(d, 0.95) ≈ 0.9
+end;