I don't know Julia. And I don't understand Bayes' Theorem. But it's interesting
that Swalwell was pressured to resign, given the accusations against Trump. So
when I originally asked Devstral to write me a julia program to estimate the
likelihood Swalwell is a harasser, it incremented α (in Β(α,β)). But decrementing
β better fits my intuition. The output seems reasonable. >8^D
Too bad Swalwell has the organs that respond to shame. He could've been
president some day, after racking up several more accusations ... and maybe a
conviction or two.
--
¡sıɹƎ ןıɐH ⊥ ɐןןǝdoɹ ǝ uǝןƃ
ὅτε oi μὲν ἄλλοι κύνες τοὺς ἐχϑροὺς δάκνουσιν, ἐγὰ δὲ τοὺς φίλους, ἵνα σώσω.
using Distributions
using Plots
# Initial parameters
initial_prob_harasser = 0.1 # Initial probability that a random person is a harasser (base rate)
initial_prob_accusation_given_harasser = 0.9 # Probability of accusation if person is a harasser
initial_prob_accusation_given_not_harasser = 0.1 # Probability of false accusation if person is not a harasser
# Prior probability that the accused is a harasser
prior_prob_harasser = initial_prob_harasser
# Function to update probability of being a harasser given a new accusation
function update_prob_harasser(prior_prob_h, prob_acc_given_h, prob_acc_given_not_h)
# P(H|A) = [P(A|H) * P(H)] / [P(A|H) * P(H) + P(A|~H) * P(~H)]
likelihood_harasser = prob_acc_given_h * prior_prob_h
likelihood_not_harasser = prob_acc_given_not_h * (1 - prior_prob_h)
posterior_prob_h = likelihood_harasser / (likelihood_harasser + likelihood_not_harasser)
return posterior_prob_h
end
# Function to plot probability and save to file
function plot_probability(probabilities, filename, title="Probability of Being a Harasser")
p = plot(1:length(probabilities), probabilities,
title=title,
xlabel="Number of Accusations",
ylabel="Probability",
label="P(H|accusations)",
ylim=(0,1),
marker=:circle)
savefig(p, filename)
println("Saved plot to $filename")
end
# Simulate receiving additional accusations
function simulate_accusations(initial_prob_h, prob_acc_given_h, prob_acc_given_not_h, num_accusations)
current_prob_h = initial_prob_h
probabilities = Float64[]
for i in 1:num_accusations
# Update probability with new accusation
current_prob_h = update_prob_harasser(current_prob_h, prob_acc_given_h, prob_acc_given_not_h)
push!(probabilities, current_prob_h)
println("After accusation $i: P(H) = $(round(current_prob_h, digits=4))")
end
return probabilities
end
# Run simulation with 18 accusations
println("Updating probability that the accused is a harasser:")
probabilities = simulate_accusations(
prior_prob_harasser,
initial_prob_accusation_given_harasser,
initial_prob_accusation_given_not_harasser,
18
)
# Plot the probability progression
plot_probability(probabilities, "harasser_probability.png").- .-.. .-.. / ..-. --- --- - . .-. ... / .- .-. . / .-- .-. --- -. --. / ...
--- -- . / .- .-. . / ..- ... . ..-. ..- .-..
FRIAM Applied Complexity Group listserv
Fridays 9a-12p Friday St. Johns Cafe / Thursdays 9a-12p Zoom
https://bit.ly/virtualfriam
to (un)subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/
archives: 5/2017 thru present https://redfish.com/pipermail/friam_redfish.com/
1/2003 thru 6/2021 http://friam.383.s1.nabble.com/
