vankesteren

library(shiny)
library(shinythemes)
library(rhandsontable)
library(shinyalert)
library(PlackettLuce)

selopts <- list(
  plugins = list("remove_button"),
  create = TRUE,
  persist = TRUE

vankesteren

import polars as pl
from metasynth import MetaFrame

# example dataframe from polars website
df = pl.DataFrame(
    {
        "ID": [1, 2, 3, 4, 5],
        "fruits": ["banana", "banana", "apple", "apple", "banana"],
        "B": [5, 4, 3, 2, 1],
        "cars": ["beetle", "audi", "beetle", "beetle", "beetle"],

vankesteren

# Implementing synthetic control using glmnet
library(tidyverse)
library(tidysynth)
library(glmnet)

# we need a wide dataset, only cigsale
smoking_w <- 
  smoking |> 
  select(state, year, cigsale) |> 
  pivot_wider(names_from = state, values_from = cigsale)

vankesteren

library(synthpop)
library(lightgbm)
library(tidyverse)

df <- SD2011

syn_res <- syn(df[,1:7])
sdf <- syn_res$syn

vankesteren

# t-mixture modeling with EM
priorp <- 0.6
m1 <- 0
m2 <- 3
s1 <- 1
s2 <- 0.707
df1 <- 2
df2 <- Inf

# generate some data with 2 classes

vankesteren

# Low-dimensional visualisation and density ratio estimation

library(tidyverse)
library(patchwork)
library(umap)
library(lfda)
library(densratio)

# generate data
N     <- 500

vankesteren

using Downloads
using CSV
using StatsKit
using Gadfly

# Apply macro from StatsBase to create new weights type
BayesianBootstrapWeights = StatsBase.@weights BayesianBoostrapWeights
function StatsBase.varcorrection(w::BayesianBootstrapWeights, corrected::Bool=false)
    s = w.sum
    if corrected

vankesteren

# Showcase
env = Schelling([0.45, 0.45, 0.1], Ba = 0.5, dim = 200)
anim = @animate for i in 1:100
    plot(env; title = i)
    step!(env)
end
gif(anim, "anim.gif", fps = 20)

vankesteren

using Distributions
using Random
using DataFrames

struct MetaVariable
    name::String
    p_missing::Float64
    dist::UnivariateDistribution
end

vankesteren

# exponential decay estimation with AR model
N <- 30
y <- numeric(N)
first_obs <- 2  # where do we start?
asymptote <- .5 # where do we go?
ar1_param <- .6 # how slow do we go there? (between -1 and 1)
sigma  <- 0

y[1] <- first_obs
for (n in 2:N) {