tvladeck

data {
  int n_respondents;
  int n_choices;
  int n_vaccines;
  int responses[n_choices];
  int respondent[n_choices];
  int vaccine[n_choices];
  vector[n_choices] money;
}

tvladeck

library(tidyverse)

set.seed(1)
sims <- map(1:200, function(x) {
  
  intercept = -runif(1)/2
  beta_male = runif(1)/4
  beta_smoker = runif(1)/4
  
  dat = 

tvladeck

set.seed(1)
margins <- c()
for(i in 1:1000) {
  sim <- c()
  for(j in 1:10) {
    experts <- rbeta(1000, 1.5, 1)
    expert_votes <- runif(1000) < experts
    uninformed <- runif(1000) < .5
    
    sim[j] = sum(expert_votes) + sum(uninformed) - sum(!expert_votes)-sum(!uninformed)

tvladeck

p_over_time = 
  read_csv("https://cdn.economistdatateam.com/us-2020-forecast/data/president/electoral_college_probability_over_time.csv") %>% 
  
  # convert win probabilites to logit scale
  mutate(win_prob_logit = boot::logit(win_prob)) %>% 
  
  # look only at one side
  filter(party == "democratic")

# take the variance of the differences

tvladeck

tvladeck

less_uncertainty_smaller_step <- 
  .25 * (1:1000 %>% map(~ run_simulation(initial = 1 - 0.1)) %>% unlist %>% mean) + 
    .5 * (1:1000 %>% map(~ run_simulation(initial = 1)) %>% unlist %>% mean) + 
    .25 * (1:1000 %>% map(~ run_simulation(initial = 1 + 0.1)) %>% unlist %>% mean) 
  
  

greater_uncertainty_bigger_step <- 
  .05 * (1:1000 %>% map(~ run_simulation(initial = 1 - 3, step_size = 0.5)) %>% unlist %>% mean) + 
  .1 * (1:1000 %>% map(~ run_simulation(initial = 1 - 2, step_size = 0.5)) %>% unlist %>% mean) + 

tvladeck

# from IPUMS
# includes 1950 forward
# includes individual income 
# sex
# age

raw_data <- data.table::fread("~/Downloads/usa_00008.csv")

library(tidyverse)

tvladeck

library(foreach)
library(doParallel)
library(tidyverse)
library(magrittr)

registerDoParallel(40)

runs <- 100

N_test <- c(250, 300, 350, 400, 500, 600) # sample in each group

tvladeck

cat orders_2018-05-01.csv | head | sed -e 's/,,/, ,/g' | column -s, -t | less -#5 -N -S

tvladeck

set.seed(20180226)
block_mining_times <- rpois(1000, 10) # lambda = 10
max_time <- sum(block_mining_times)
cumulative_times <- cumsum(block_mining_times)

rand_times <- runif(1000, min = 0, max = max_time)

deltas_between_blocks <- 
  rand_times %>% 
  sapply(function(t){