JosepER

Mode <- function(x, tie = c("missing", "error"), ignore_categories = NULL) {
  
  if(!is.null(ignore_categories)){

        x <- x[!(x %in% ignore_categories)]

  }
  
  

JosepER

# produce balance statistics -------


    ## version to pass unquoted
balance_stats_var <- function(data_, treatment_, variable_){
  
  treatment_ <- enquo(treatment_)
  variable_ <- enquo(variable_)
  
  temp_means <- group_by(data_, !!treatment_) %>% ## make it possible to change the 'treatment' var

JosepER

#include <Rcpp.h>
using namespace Rcpp;

// [[Rcpp::export]]
IntegerVector recode_outcomes(IntegerVector x) {
  
  int n = x.size();
  IntegerVector y(n);
  
  for(int i=0; i<n; i++){

JosepER

#' Compute gini index.
#' 
#' Compute the gini index with (optional) weights.
#' 
#' @param x A numeric vector with the data.
#' @param weights A numeric vector with sample weights.
#' @return res An atomic vector with the gini index.
gini <- function(x, weights = NULL, na.rm = FALSE) {
  

JosepER

# Prepare session 
 
library(rlang) 
library(glue) 
library(fs) 
library(dplyr) 
library(magrittr) 
library(purrr) 
library(ggplot2) 
 

JosepER

library(readr) 
library(dplyr)   
library(magrittr)   
library(purrr)   
library(ggplot2)   
 
all_lissyrtools_scripts <- fs::dir_ls("/media/user/lissyrtools/")   
 
invisible(purrr::map(all_lissyrtools_scripts, ~ source(.x)))   
 

JosepER

import re

def backward_string_by_word(text: str) -> str:

    spaces = re.findall(" +", text)

    words = text.split()


    if len(words) == 0:

JosepER

#' Weighted quantiles
#'
#' Wrap over Julia's 'StatsBase.quantile()' function.
#'
#' @param x A numeric vector for which the quantiles should be computed.
#' @param weights A numeric vector with the weights.
#' @param probs A numeric vector with the quantiles to compute.
#' @param na.rm a logical value indicating whether NA values should be stripped before the computation proceeds.
#'
#' @return A numeric vector with the weighted quantiles.

JosepER

##### Atkinson index #####
"""
    atkinson(v, w, ϵ)
Computes the weighted Atkinson index given an inequality adversion parameter ϵ.
"""
function wtd_atkinson(v::Array{<:Real,1}, w::Array{<:Real,1}, ϵ::Real) ::Float64
 
    v = v/Statistics.mean(v)
    w = w/sum(w)
 

JosepER

weighted_atkinson_jl <- function(x, weights, epsilon, na.rm = FALSE){
  
  JuliaCall::julia_call("weighted_atkinson", x, weights, epsilon, na.rm)
  
}

julia_weighted_atkinson_function <- JuliaCall::julia_eval("function weighted_atkinson(v, w, ϵ::Real, skipmissing::Bool)
    if skipmissing
           w = w[findall(!ismissing, v)]
           v = v[findall(!ismissing, v)]