steveharoz

library(tidyverse)
library(lmerTest)

# subject count
COUNT = 5

set.seed(8675309)

# generate a unique intercept per subject
data = tibble(

steveharoz

steveharoz

# Extract all citation numbers such as [1] from a PDF's text
# It also includes cases for multiples [1, 3] and ranges [1-5]
# It tries to exclude confidence intervals by skipping 
#
# written by Steve Haroz with help from ChatGPT
# MIT license

library(tidyverse)
library(pdftools)

steveharoz

COUNT = 100000

# How often does a single t-test of random data yield p<0.05?
replicate(COUNT, 
  t.test(rnorm(20))$p.value < 0.05
) %>% mean()

#> 0.05073
# 5% false positive rate

steveharoz

library(tidyverse)

COUNT = 40

data = tibble(
  car = paste0(sample(LETTERS, COUNT, TRUE), sample(letters, COUNT, TRUE), sample(letters, COUNT, TRUE)),
  value = rnorm(COUNT, 3),
  group = c(rep("Petrol", COUNT/2), rep("Hybrid", COUNT/4), rep("Pure Electric", COUNT/8), rep("Diesel", COUNT/8))
)

steveharoz

StatEndpoint <- ggproto("StatEndpoint", Stat,
  compute_group = function(data, scales) {
    # sort by x so indexing is meaningful
    data = arrange(data, x)
    # grab only the first and last row
    data[c(1,nrow(data)),]
  },
  
  required_aes = c("x", "y")
)

steveharoz

x,y,value
1,205,0.3125
1,204,0.3125
1,203,0.3125
1,202,0.3125
1,201,0.3125
1,200,0.3125
1,199,0.3125
1,198,0.3125
1,197,0.3125

steveharoz

painbow moved to github: https://github.com/steveharoz/painbow

steveharoz

library(tidyverse)

set.seed(999)

data = tibble(
  name = c("A1", "A2", "A3", "A4", "B1", "B2", "B3", "B4", "C1", "C2"),
  value = rnorm(10, 10, sd = 3),
  color = c(
    hcl(220, seq(60, 30, -10), seq(50, 80, 10)),
    hcl(0, seq(60, 30, -10), seq(50, 80, 10)),

steveharoz

library(tidyverse)

# arbitrary number
district_count = 38

# population from stephanie's figure
# https://twitter.com/evergreendata/status/1450862060972216320
population = c(
  rep("White", 40),
  rep("Latino", 39),