JimGrange

# step 1 (before)
tmp <- installed.packages()
installedpackages <- as.vector(tmp[is.na(tmp[,"Priority"]), 1])
save(installedpackages, file="installed_packages.rda")

# now install new version of R

# step 2 (after)
load("installed_packages.rda")
for(i in 1:length(installedpackages)){

JimGrange

# generate table of required sample sizes for various correlation effect sizes
# and various levels of desired power in a study

# you need the pwr package
library(pwr)

# vector of effect sizes to explore
effect_sizes <- seq(from = 0.05, to = 0.95, by = 0.05)

# vector of desired power

JimGrange

JimGrange

#------------------------------------------------------------------------------
### Packages & functions

library(dplyr)
library(afex)
library(here)

# generate data from a multivariate normal distribution with known 
# correlations between values
mvrnorm <- function(n, nValues, meanValues, sdValues, corMatrix) {

JimGrange

library(pwr)
library(effsize)
library(ggplot2)

# set "true" effect size and the effect size entered into power analysis
true_effect <- 0.2
power_effect <- 1.0

# How many participants required?
power_analysis <- pwr.t.test(d = power_effect, 

JimGrange

# iterate over starting parameters and assess density-----------------

do_fit <- function(current_data){

  # mu <- seq(from = 0.2, to = 1.8, by = 0.2)
  # sigma <- seq(from = 0.05, to = 0.65, by = 0.2)
  # tau <- seq(from = 0.05, to = 0.85, by = 0.2)
  
  mu <- 0.5
  sigma <- 0.05

JimGrange

# sample size
n <- 500

# simulate data for each predictor 
# (note predictor b is categorical)
predictor_a <- rnorm(n, 0, 1) 
predictor_b <- rbinom(n, 1, .5)

# population-level beta values
b_predictor_a <- -0.20

JimGrange

library(tidyverse)
library(rstan)
rstan_options(auto_write = TRUE)

#--- declare the data
# n_c = total number of people who drink caffeine
# n_nc = total number of people who do not drink caffeine
# p_c = observed proportion of people who drink caffeine that have a favourite mug
# p_nc = observed proportion of people who drink caffeine that DO NOT have a favourite mug
stan_data <- list(

JimGrange

mod_1 <- brm(value ~ questionnaire * condition, 
             data = idealised_data,
             seed = 42,
             cores = 4)

mod_2 <- brm(value ~ questionnaire + condition, 
             data = idealised_data,
             seed = 42,
             cores = 4)

JimGrange

# gaussian kernel function
gaussian_kernel <- function(u){
  (1 / sqrt(2 * pi)) * exp(-0.5 * u ^ 2)
}


# kernel density estimate function
kde <- function(n, data, x_limit, y_limit, h_x, h_y){
  
  x <- seq(from = x_limit[1],