fix loo_compare in docs, simplify turing model, update project.toml

docs/Project.toml

@@ -5,6 +5,7 @@ Distributions = "31c24e10-a181-5473-b8eb-7969acd0382f"
 Documenter = "e30172f5-a6a5-5a46-863b-614d45cd2de4"
 FillArrays = "1a297f60-69ca-5386-bcde-b61e274b549b"
 KernelDensity = "5ab0869b-81aa-558d-bb23-cbf5423bbe9b"
+ParetoSmooth = "a68b5a21-f429-434e-8bfa-46b447300aac"
 Plots = "91a5bcdd-55d7-5caf-9e0b-520d859cae80"
 Random = "9a3f8284-a2c9-5f02-9a11-845980a1fd5c"
 Revise = "295af30f-e4ad-537b-8983-00126c2a3abe"
@@ -18,6 +19,7 @@ Colors = "0.12.0"
 DataFrames = "1.0.0"
 Documenter = "1"
 KernelDensity = "0.6.0"
+ParetoSmooth = "0.7.0"
 Plots = "1.0.0"
 StatsBase = "0.33.0,0.34.0"
 StatsModels = "0.7.0"

docs/src/loo_compare.md

@@ -10,20 +10,19 @@ First, we'll simulate data from a [`LBA`](@ref) with fixed parameters. Then, we'
 
 Before proceeding, we will load the required packages.
 
-```@setup loo_compare
+```@example loo_example
 using Random
 using SequentialSamplingModels
 using LinearAlgebra
 using ParetoSmooth
 using Turing
-
 ```
 
 ## Data-Generating Model
 
 The next step is to generate simulated data for comparing the models. Here, we'll use an LBA as the true data-generating model:
 
-```@example loo_compare
+```@example loo_example
 Random.seed!(5000)
 
 dist = LBA(ν=[3.0, 2.0], A=0.8, k=0.3, τ=0.3)
@@ -34,52 +33,46 @@ data = rand(dist, 1000)
 
 The following code block defines the model along with its prior distributions using Turing.jl. We'll use this model with different fixed values for the `k` parameter.
 
-```@example loo_compare
-@model function model_LBA(data, k; min_rt=0.2)
+```@example loo_example
+@model function model_LBA(data, k; min_rt=minimum(data.rt))
     # Priors
     ν ~ MvNormal(fill(3.0, 2), I * 2)
     A ~ truncated(Normal(0.8, 0.4), 0.0, Inf)
     τ ~ Uniform(0.0, min_rt)
 
     # Likelihood
-    for i in 1:length(data)
-        data[i] ~ LBA(; ν, A, k, τ)
-    end
+    data ~ LBA(; ν, A, k, τ)
 end
 ```
 
 ## Estimate the Parameters
 
 Now we'll estimate the parameters using three different fixed `k` values:
 
-```@example loo_compare
-#prepare the data for model fitting
-dat = [(choice=data.choice[i], rt=data.rt[i]) for i in 1:length(data.rt)]
-min_rt = minimum(data.rt)
-
-chain_LBA1 = sample(model_LBA(dat, 2,  min_rt=min_rt), NUTS(), 1000)
-chain_LBA2 = sample(model_LBA(dat, 0.3,min_rt=min_rt), NUTS(), 1000)
-chain_LBA3 = sample(model_LBA(dat, 1,  min_rt=min_rt), NUTS(), 1000)
+```@example loo_example
+chain_LBA1 = sample(model_LBA(data, 2.0), NUTS(), 1000)
+chain_LBA2 = sample(model_LBA(data, 0.3), NUTS(), 1000)
+chain_LBA3 = sample(model_LBA(data, 1.0), NUTS(), 1000)
 ```
 
 ## Compute PSIS-LOO
 
 Next we will use the `psis_loo` function to compute the PSIS-LOO for each model:
 
-```@example loo_compare
-res1 = psis_loo(model_LBA(dat, 2,   min_rt=min_rt), chain_LBA1)
-res2 = psis_loo(model_LBA(dat, 0.3, min_rt=min_rt), chain_LBA2)
-res3 = psis_loo(model_LBA(dat, 1, min_rt=min_rt), chain_LBA3)
-show(stdout, MIME"text/plain"(), ans) #hide
+```@example loo_example
+res1 = psis_loo(model_LBA(data, 2.0), chain_LBA1)
+res2 = psis_loo(model_LBA(data, 0.3), chain_LBA2)
+res3 = psis_loo(model_LBA(data, 1.0), chain_LBA3)
+show(stdout, MIME"text/plain"(), ans)
 ```
 
 ## Compare Models
 
 Finally, we can compare the models using the `loo_compare` function:
 
-```@example loo_compare
-loo_compare((LBA1=res1, LBA2=res2, LBA3 = res3))
-show(stdout, MIME"text/plain"(), ans) # hide
+```@example loo_example
+loo_compare((LBA1 = res1, LBA2 = res2, LBA3 = res3))
+show(stdout, MIME"text/plain"(), ans)
 ```
 
 Here we indeed correctly identified the generative model we simulated. It is of note that some researchers have criticized using model comparison metrics such as leave-one-out cross-validation. See Gronau et al. (2019) for more information.