turgeonmaxime

library(stringr)

data_path <- 'somefolder/planet04_rgbn_c9_r8.txt'

# We want to extract 9 and 8
basename(data_path)

# Too many numbers
str_extract_all(basename(data_path), "([0-9]+)")
# [[1]]

turgeonmaxime

library(tidyverse)

expit <- function(t) exp(t)/(1 + exp(t))

n <- 1000000
prev_vec <- c(0.01, 0.05, 0.1, 0.25, 0.5)

results <- purrr::map_df(prev_vec, \(prev) {
    # Generate data
    dvec <- rbinom(n, prob = prev, size = 1)

turgeonmaxime

library(tidyverse)
library(cowplot)

gg1 <- ggplot(ToothGrowth, aes(x = len)) +
    geom_dotplot(aes(fill = as.factor(dose)),
                 binwidth = .5) +
    guides(fill = FALSE)

ToothGrowth2 <- arrange(ToothGrowth, len)
gg2 <- ggplot(ToothGrowth, aes(x = len, group = factor(dose))) +

turgeonmaxime

library(tidyverse)
library(nycflights13)

delay <- flights |> 
    group_by(dest) |> 
    summarize(means = mean(dep_delay, na.rm = TRUE))

joined <- inner_join(delay, flights, by = "dest")

turgeonmaxime

#----
# Poisson vs casebase
# authors: Max Turgeon, Jesse Islam and Sahir Bhatnagar
# date: 10/1/2021
#----
set.seed(1952)

library(casebase)
library(cowplot)
library(Epi)

turgeonmaxime

B <- 1000
n <- 20
sigma <- 10
p <- 0.9
alpha <- 0.05

results <- replicate(B, {
    norm_vars1 <- rnorm(n)
 # Contaminated normal

turgeonmaxime

library(casebase)
library(survival)
library(splines)
library(tidyverse)
library(cowplot)

# 1. Fit casebase with splines----
data("ERSPC")
ERSPC <- mutate(ERSPC, ScrArm = factor(ScrArm, 
                                       levels = c(0,1), 

turgeonmaxime

library(tidyverse)
library(RcppRoll)
library(rvest)


# Create temporary directory
tmp_dir <- tempdir()
file_path <- paste0(tmp_dir, "/vaccine_sk.csv")

# Download file

turgeonmaxime

library(tidyverse)
library(lubridate)
library(readr)

data_hr <- read_csv("https://raw.githubusercontent.com/ishaberry/Covid19Canada/master/timeseries_hr/cases_timeseries_hr.csv")

# Coerce string to dates and factor provinces
data_cum <- data_hr %>%
    filter(province %in% c("Alberta", "Saskatchewan")) %>% 
    mutate(date_report = lubridate::dmy(date_report)) %>%

turgeonmaxime

library(tidyverse)
library(lubridate)
library(readr)
library(ggbeeswarm)

# Download data on cases by health region
data_hr <- read_csv("https://raw.githubusercontent.com/ishaberry/Covid19Canada/master/timeseries_hr/cases_timeseries_hr.csv")

# Filter for province, coerce string to dates, and add variable
# to highlight certain regions