add LNR model plot

Author: itsdfish

Project.toml

@@ -1,7 +1,7 @@
 name = "SequentialSamplingModels"
 uuid = "0e71a2a6-2b30-4447-8742-d083a85e82d1"
 authors = ["itsdfish"]
-version = "0.11.10"
+version = "0.11.11"
 
 [deps]
 Distributions = "31c24e10-a181-5473-b8eb-7969acd0382f"

docs/src/bayes_factor.md

@@ -131,7 +131,7 @@ mll_rdm = stepping_stone(pt_rdm)
 ```
 
 ## Compute the Bayes Factor
-The bayes factor is obtained by exponentiating the difference between marginal log likelihoods. The value of `1.21` indicates that the LBA is `1.21` times more likely to have generated the data. 
+The bayes factor is obtained by exponentiating the difference between marginal log likelihoods. The value of `1.21` indicates that the data are `1.21` times more likely under the LBA than the RDM. 
 ```julia
 bf = exp(mll_lba - mll_rdm)
 ```

ext/plots/plot_model.jl

@@ -119,7 +119,7 @@ end
 compute_threshold(model) = model.α
 compute_threshold(model::AbstractLBA) = model.A + model.k
 compute_threshold(model::AbstractRDM) = model.A + model.k
-
+compute_threshold(model::AbstractLNR) = 1.0
 """
     get_default_labels(model::AbstractRDM)
 
@@ -153,6 +153,23 @@ function get_default_labels(model::AbstractLBA)
     ]
 end
 
+"""
+    get_default_labels(model::AbstractLNR)
+
+Generates default parameter labels and locations 
+
+# Arguments
+
+- `model::AbstractLBA`: an object for the log normal race model 
+"""
+function get_default_labels(model::AbstractLNR)
+    α = 1
+    return [
+        (0, α, text("α", 10, :right)),
+        (model.τ, 0, text("τ", 10, :bottom))
+    ]
+end
+
 """
     get_default_labels(model::AbstractRDM)
 
@@ -483,6 +500,34 @@ function get_model_plot_defaults(d::AbstractLBA)
     )
 end
 
+"""
+    get_model_plot_defaults(d::AbstractLNR)
+
+Returns default plot options 
+
+# Arguments
+
+- `d::AbstractLNR`: an object for the log normal race 
+- `n_subplots`: the number of subplots (i.e., choices)
+"""
+function get_model_plot_defaults(d::AbstractLNR)
+    n_subplots = n_options(d)
+    title = ["choice $i" for _ ∈ 1:1, i ∈ 1:n_subplots]
+    return (
+        xaxis = nothing,
+        yaxis = nothing,
+        xticks = nothing,
+        yticks = nothing,
+        grid = false,
+        linewidth = 0.75,
+        color = :black,
+        leg = false,
+        title,
+        layout = (n_subplots, 1),
+        arrow = :closed
+    )
+end
+
 """
     get_model_plot_defaults(d::AbstractCDDM)
 

src/LNR.jl

@@ -109,3 +109,29 @@ function pdf(d::AbstractLNR{T, T1}, r::Int, t::Float64) where {T, T1 <: Real}
     end
     return density
 end
+
+"""
+    simulate(model::AbstractLNR; n_steps=100, _...)
+
+Returns a matrix containing evidence samples of the LBA decision process. In the matrix, rows 
+represent samples of evidence per time step and columns represent different accumulators.
+
+# Arguments
+
+- `model::AbstractLNR`: a subtype of AbstractLNR
+
+# Keywords 
+
+- `n_steps=100`: number of time steps at which evidence is recorded
+"""
+function simulate(rng::AbstractRNG, model::AbstractLNR; n_steps = 100, _...)
+    (; τ, ν, σ) = model
+    n = length(ν)
+    νs = @. rand(rng, Normal(ν, σ))
+    βs = @. exp(νs)
+    _,choice = findmax(βs)
+    t = 1 / βs[choice]
+    evidence = collect.(range.(0, βs * t, length = 100))
+    time_steps = range(0, t, length = n_steps)
+    return time_steps, hcat(evidence...)
+end

test/plots.jl

@@ -242,6 +242,30 @@
         plot_model(dist; add_density = true, n_sim = 2, density_kwargs, xlims = (0, 1.2))
     end
 
+    @safetestset "LNR" begin
+        using Plots
+        using SequentialSamplingModels
+        using Test
+
+        dist = LNR()
+
+        h = histogram(dist)
+        plot!(h, dist)
+
+        histogram(dist)
+        plot!(dist)
+
+        p = plot(dist)
+        histogram!(p, dist)
+
+        plot(dist)
+        histogram!(dist)
+
+        density_kwargs = (; t_range = range(0.1, 1.2, length = 100),)
+        plot_model(dist; add_density = true, n_sim = 2, density_kwargs, xlims = (0, 1.2))
+    end
+
+
     @safetestset "WaldMixture" begin
         using Plots
         using SequentialSamplingModels