vankesteren’s gists

vankesteren / diffiehellman.py

Last active October 13, 2025 07:36

Diffie-Hellman key exchange

	from random import randint


	class Agent:
	"""Agent that can share their public key with another agent to compute a shared secret, without sharing that secret."""

	def __init__(self, name: str, prime: int = 217 - 1, base: int = 25 - 1) -> None:
	self.name = name
	self.prime = prime
	self.base = base

vankesteren / tidysim.py

Last active July 19, 2025 18:59

Tidy simulation example in python

	import polars as pl
	import numpy as np
	from polarsgrid import expand_grid
	from scipy.stats import norm, t, uniform, ttest_ind
	from tqdm import tqdm
	import plotnine as p9

	grid = expand_grid(
	# data generating process parameters
	sample_size=[10, 20, 40, 80, 160, 320, 640],

vankesteren / causal_graph_sim.R

Created October 31, 2024 12:58

Simulating data, estimating effects, and computing intervention effects with uncertainty propagation in a simple causal graph

	# Packages for this script
	library(tidyverse)
	library(lavaan)
	library(dagitty)
	library(glue)
	library(rstanarm)


	## SIMULATION ##
	# here is a true data-generating process in lavaan (SEM) syntax

vankesteren / virtualgrid.R

Last active September 6, 2024 19:25

Stop storing huge full-factorial grids, start using virtual grids. It behaves like a data frame but uses only a fraction of the memory!


	virtual_grid <- function(...) {
	pars <- list(...)
	lens <- vapply(pars, length, 1L)
	return(structure(list(pars = pars, lens = lens), class = c("vgrid", "tbl_lazy")))
	}

	dim.vgrid <- function(x) {
	as.integer(c(prod(x$lens), length(x$lens)))
	}

vankesteren / scm_factor.R

Created August 22, 2024 15:05

SCM factor structure (WIP)

	# panel data factor model simulation
	# See hollingsworth & wing (2022) tactics for design..
	# paragraph 2.3, assumption 2
	n_timepoints <- 100
	n_unobserved <- 1
	n_units <- 200
	n_covar <- 5

	ru <- 0.6
	mean_ly <- 1

vankesteren / importance_sampling_vs_lintsampler.py

Last active June 26, 2024 18:25

Comparing lintsampler to basic uniform importance sampling

	# Comparing lintsampler to basic uniform importance sampling
	from scipy.stats import norm, uniform
	import numpy as np
	import matplotlib.pyplot as plt
	from lintsampler import LintSampler

	NSAMPLES = 1000000

	# GMM example
	def gmm_pdf(x):

vankesteren / probsocsim.R

Created June 19, 2024 10:41

Example of a probabilistic social simulation

	# Simple probabilistic simulation script

	# Causal graph:
	# NetIncome -> + CulturalActivities
	# NetIncome -> + SportsActivities
	# NetIncome -> - Debts
	# NetIncome -> + SocialComparison
	# SportsActivities -> + Health
	# SportsActivities -> + Partnership
	# CulturalActivities -> + SocialComparison

vankesteren / permutefun.jl

Last active September 6, 2024 08:31

Permuting data to induce complex bivariate relations.

	using StatsBase: sample, mean, cor
	using LinearAlgebra: norm
	using Plots, Random

	"""
	permutefun!(x::Vector, y::Vector, rule::Function, score::Real; tol::Number = 1e-3, max_iter::Int = 10_000, max_search::Number = 100, verbose::Bool = true)

	Permute y values to approximate a functional constraint (rule) between x and y.

	# Arguments

vankesteren / rol_model.jl

Last active May 10, 2024 10:13

Bayesian inference for rank-ordered logit model in Julia's Turing.jl

	using Turing
	using LogExpFunctions: logsumexp
	using DataFrames

	# The rank ordered logit model in Turing
	# The rank-ordered logit likelihood
	function rank_ordered_logit(ordered_skills::Vector{<:Real})
	ll = 0.0
	for m in 1:(length(ordered_skills) - 1)
	ll += ordered_skills[m] - logsumexp(ordered_skills[m:end])

vankesteren / idealpoint.R

Last active April 25, 2024 19:55

Ideal point model in stan

	# Ideal point model in stan / R
	# example taken & simplified from https://medewitt.github.io/resources/stan_ideal_point.html
	library(tidyverse)
	library(cmdstanr)

	# simulate data: 100 legislators, 150 votes
	set.seed(1834)
	N_legislators <- 50
	N_bills <- 150

Erik-Jan van Kesteren vankesteren