add DDM tests and adjust DDM plot docs

docs/make.jl

@@ -16,6 +16,7 @@ makedocs(
     pages = ["Home" => "index.md",
             "API" => "api.md",
             "Models" => ["Attentional Drift Diffusion" => "aDDM.md",
+                "Drift Diffusion Model" => "DDM.md",
                 "Leaky Competing Accumulator" => "lca.md",
                 "Linear Ballistic Accumulator" => "lba.md",
                 "Lognormal Race Model" => "lnr.md",

docs/src/DDM.md

@@ -52,7 +52,7 @@ The duration for a non-decisional processes (encoding and response execution). T
 An indicator of an an initial bias towards a decision. The z parameter is relative to a (i.e. it ranges from 0 to 1).
 
 ```@example DDM 
-z = 0.25
+z = 0.50
 ```
 
 ### DDM Constructor 
@@ -94,14 +94,14 @@ rts1 = rts[choices .== 1]
 rts2 = rts[choices .== 2]
 # probability of choosing 1
 p1 = length(rts1) / length(rts)
-t_range = range(.30, 2, length=100)
+t_range = range(.30, 1, length=100)
 # pdf for choice 1
 pdf1 = pdf.(dist, (1,), t_range)
 # pdf for choice 2
 pdf2 = pdf.(dist, (2,), t_range)
 # histogram of retrieval times
 hist = histogram(layout=(2,1), leg=false, grid=false,
-     xlabel="Reaction Time", ylabel="Density", xlims = (0,1.5))
+     xlabel="Reaction Time", ylabel="Density", xlims = (0,1.))
 histogram!(rts1, subplot=1, color=:grey, bins = 200, norm=true, title="Choice 1")
 plot!(t_range, pdf1, subplot=1, color=:darkorange, linewidth=2)
 histogram!(rts2, subplot=2, color=:grey, bins = 150, norm=true, title="Choice 2")

test/ddm_tests.jl

@@ -0,0 +1,101 @@
+@safetestset "DDM Tests" begin
+    @safetestset "DDM pdf 1" begin
+        using SequentialSamplingModels
+        using Test
+        using KernelDensity
+        using Random
+        Random.seed!(654)
+
+        dist = DDM(ν=1.0, α = .8, z = .5, τ = .3) 
+        choice,rt = rand(dist, 10^5)
+        rt1 = rt[choice .== 1]
+        p1 = mean(choice .== 1)
+        p2 = 1 - p1
+        approx_pdf = kde(rt1)
+        x = range(.301, 1.5, length=100)
+        y′ = pdf(approx_pdf, x) * p1
+        y = pdf.(dist, (1,), x)
+        @test y′ ≈ y rtol = .10
+
+        rt2 = rt[choice .== 2]
+        approx_pdf = kde(rt2)
+        y′ = pdf(approx_pdf, x) * p2
+        y = pdf.(dist, (2,), x)
+        @test y′ ≈ y rtol = .10
+    end
+
+    @safetestset "DDM pdf 2" begin
+        using SequentialSamplingModels
+        using Test
+        using KernelDensity
+        using Random
+        Random.seed!(750)
+
+        dist = DDM(ν=2.0, α = 1.5, z = .5, τ = .30) 
+        choice,rt = rand(dist, 10^5)
+        rt1 = rt[choice .== 1]
+        p1 = mean(choice .== 1)
+        p2 = 1 - p1
+        approx_pdf = kde(rt1)
+        x = range(.301, 1.5, length=100)
+        y′ = pdf(approx_pdf, x) * p1
+        y = pdf.(dist, (1,), x)
+        @test y′ ≈ y rtol = .05
+
+        rt2 = rt[choice .== 2]
+        approx_pdf = kde(rt2)
+        y′ = pdf(approx_pdf, x) * p2
+        y = pdf.(dist, (2,), x)
+        @test y′ ≈ y rtol = .10
+    end
+
+    @safetestset "DDM cdf 1" begin
+        using SequentialSamplingModels
+        using Test
+        using StatsBase
+        using Random
+        Random.seed!(7540)
+
+        dist = DDM(ν=1.0, α = .8, z = .5, τ = .3) 
+        choice,rt = rand(dist, 10^5)
+        rt1 = rt[choice .== 1]
+        p1 = mean(choice .== 1)
+        p2 = 1 - p1
+        ecdf1 = ecdf(rt1)
+        x = range(.31, 1.0, length=100)
+        y′ = ecdf1.(x) * p1
+        y = cdf.(dist, (1,), x)
+        @test y′ ≈ y rtol = .01
+
+        rt2 = rt[choice .== 2]
+        ecdf2 = ecdf(rt2)
+        y′ = ecdf1.(x) * p2
+        y = cdf.(dist, (2,), x)
+        @test y′ ≈ y rtol = .01
+    end
+
+    @safetestset "DDM cdf 2" begin
+        using SequentialSamplingModels
+        using Test
+        using StatsBase
+        using Random
+        Random.seed!(2200)
+
+        dist = DDM(ν=2.0, α = 1.5, z = .5, τ = .30) 
+        choice,rt = rand(dist, 10^5)
+        rt1 = rt[choice .== 1]
+        p1 = mean(choice .== 1)
+        p2 = 1 - p1
+        ecdf1 = ecdf(rt1)
+        x = range(.31, 1.0, length=100)
+        y′ = ecdf1.(x) * p1
+        y = cdf.(dist, (1,), x)
+        @test y′ ≈ y rtol = .01
+
+        rt2 = rt[choice .== 2]
+        ecdf2 = ecdf(rt2)
+        y′ = ecdf1.(x) * p2
+        y = cdf.(dist, (2,), x)
+        @test y′ ≈ y rtol = .01
+    end
+end

test/distributions_api.jl

@@ -26,6 +26,7 @@
             LNR(μ=[-1.0,3.0], σ=.3, ϕ=0.0),
             DiffusionRace(;ν=[1.0,.5], k=0.5, A=1.0, θ=.2),
             Wald(3,1,.2),
+            DDM(),
             WaldMixture(2, .2, 1, .1),
             aDDM(),
             maaDDM()]
@@ -45,6 +46,7 @@
             LNR(μ=[-1.0,3.0], σ=.3, ϕ=0.0),
             DiffusionRace(;ν=[1.0,.5], k=0.5, A=1.0, θ=.2),
             Wald(3,1,.2),
+            DDM(),
             WaldMixture(2, .2, 1, .1),
             aDDM(),
             maaDDM()]