chrishanretty

### Load libraries
library(tidyverse)
library(haven) ## for Stata import
library(readxl)
library(marginaleffects) ## for easy predictions and contrasts
library(nnet) ## for nnet::multinom
library(furrr) ## to perform computations in parallel

### Set fewer workers if you have fewer cores
plan(multisession, workers = 12)

chrishanretty

### PURPOSE OF THIS CODE: estimate a large model in bam and speed-test
### predictions using mgcv::predict.bam and
### marginaleffects::predictions with and without discretization and
### parallelization

### ##################################################################
### load libraries
### ##################################################################
library(mgcv)
library(marginaleffects)

chrishanretty

library(tidyverse)
library(rstan)
library(cmdstanr)
dat <- read.delim("data_attractiveness_juge_results.csv")

### Set subject IDs using the same factor levels
subject_levels <- unique(c(dat$ID_left,
                           dat$ID_right))

dat <- dat |>

chrishanretty

library(tidyverse)
library(haven)

### Replace your file location here
bes <- read_dta("~/Dropbox/bes_data/BES2019_W25_v25.0.dta")

### What was turnout like in the panel?
bes |>
    filter(!is.na(localTurnoutRetro)) |>
    group_by(localTurnoutRetro) |>

chrishanretty

library(tidyverse)

### Write a function to download the data from the IPU
dl_data <- function(y, m) {
    url <- paste0("https://data.ipu.org/api/women-ranking.csv?load-entity-refs=taxonomy_term%2Cfield_collection_item&max-depth=2&langcode=en&month=",
                  m,
                  "&year=",
                  y)
    download.file(url, destfile = paste0("ipu_", y, m, ".csv"))
}

chrishanretty

### Load the necessary libraries
library(sf)
library(tidyverse)

### Read in the shapefile
### Downloaded from https://geoportal.statistics.gov.uk/datasets/ons::local-authority-districts-december-2022-boundaries-uk-bfc/explore?location=55.215430%2C-3.316939%2C6.61
lad_shp <- st_read("LAD_DEC_2022_UK_BFC.shp")

### Read in the list of coordinates
### 

chrishanretty

library(vdemdata)
library(tidyverse)
library(ggflags)
library(countrycode)
library(hrbrthemes)
### Get the V-Dem data in
data("vdem")
dat <- vdem |>
    distinct(country_name, country_text_id, year,
             v2x_polyarchy, v2x_polyarchy_sd) |>

chrishanretty

library(tidyverse)
library(hrbrthemes)

### This section reads in the data -- scroll down for the code that
### generates the figures and the graphics


tibble::tribble(
  ~aye_vote, ~ONSConstID,                             ~Constituency,                          ~MP,             ~Party, ~ConstyOpposition, ~ConstyOpposition_sd,
          1, "E14000530",                               "Aldershot",               "Leo Docherty",     "Conservative",  1.67320394423204,   0.0170138236227471,

chrishanretty

### Purpose of this code: plot the proportion of people in each CSES
### wave 5 study who can place themselves on a left-right scale.

### You will need to have downloaded the CSES 5 data somewhere
library(tidyverse)
library(ggtext)
library(RColorBrewer)
library(countrycode)
library(hrbrthemes)

chrishanretty

library(tidyverse)
library(hrbrthemes)
library(ggtext)

dat <- read.csv("view_cabinet.csv")

dat <- dat %>%
    group_by(country_name_short,
             country_id,
             election_date,