using StatisticalRethinking
using StatsFuns #logistic() function
using Turing

Turing.setadbackend(:reverse_diff)

d = CSV.read(joinpath(dirname(Base.pathof(StatisticalRethinking)), "..", "data",
    "chimpanzees.csv"), delim=';')
size(d) # Should be 504x8

(504, 8)

pulledleft, actors, condition, prosocleft

@model m10_4(y, actors, x₁, x₂) = begin

Number of unique actors in the data set

    N_actor = length(unique(actors))

Set an TArray for the priors/param

    α = TArray{Any}(undef, N_actor)

For each actor [1,..,7] set a prior

    for i ∈ 1:length(α)
        α[i] ~ Normal(0,10)
    end

    βp ~ Normal(0, 10)
    βpC ~ Normal(0, 10)

    for i ∈ 1:length(y)
        p = logistic(α[actors[i]] + (βp + βpC * x₁[i]) * x₂[i])
        y[i] ~ Binomial(1, p)
    end
end

posterior = sample(m10_4(d[:,:pulled_left], d[:,:actor],d[:,:condition],
    d[:,:prosoc_left]),
    Turing.NUTS(2500, 500, 0.95))
describe(posterior)

         Mean           SD       Naive SE       MCSE         ESS

βpC -0.181394034 0.28946513 0.0057893027 0.0266018473 118.404696 α[1] -0.750003835 0.29794860 0.0059589721 0.0278564821 114.401093 α[2] 9.372324757 4.49534602 0.0899069204 0.8306772626 29.286104 α[3] -1.067303557 0.32767901 0.0065535802 0.0329610836 98.831248 α[4] -1.060339791 0.27627393 0.0055254786 0.0298318788 85.766681 α[5] -0.707040908 0.28169208 0.0056338417 0.0226231747 155.039440 α[6] 0.218884973 0.27572737 0.0055145474 0.0312307275 77.946310 α[7] 1.803365110 0.38459001 0.0076918002 0.0392945213 95.792607 βp 0.870271237 0.27987766 0.0055975532 0.0324186869 74.532489

Rethinking mean sd 5.5% 94.5% n_eff Rhat a[1] -0.74 0.27 -1.17 -0.31 3838 1 a[2] 11.02 5.53 4.46 21.27 1759 1 a[3] -1.05 0.28 -1.50 -0.61 3784 1 a[4] -1.05 0.27 -1.48 -0.62 3761 1 a[5] -0.74 0.27 -1.18 -0.32 4347 1 a[6] 0.21 0.27 -0.23 0.66 3932 1 a[7] 1.81 0.39 1.19 2.46 4791 1 bp 0.84 0.26 0.42 1.26 2586 1 bpC -0.13 0.30 -0.63 0.34 3508 1

This page was generated using Literate.jl.