ito4303

##
## Chapter 4 - Exercises
##

library(cmdstanr)
options(mc.cores = 4)

# Directories

orig_dir <- "multiple_choice"

ito4303

##
## 4.4 Performing a simulation experiment to understand the model
##

library(cmdstanr)
library(ggplot2)
library(ggmcmc)
library(patchwork)
library(stringr)
library(bayesplot)

ito4303

#include <TMB.hpp>

template<class Type>
Type objective_function<Type>::operator() ()
{
  DATA_INTEGER(N); // Num. participants (N >= 0)
  DATA_INTEGER(W); // Num. participants claim the prize (0 <= W <= N)
  DATA_SCALAR(q);  // Prob. win the prize (0 <= q <= 1)
  PARAMETER(p);    // Prob. honest (0 <= p <= 1)
  Type nll;        // Negative log likelihood

ito4303

library(cmdstanr)
library(posterior)
library(ggplot2)

model_file <- file.path("models", "exercise_3.2.stan")

stan_data <- list(N = 171, W = 111, q = 0.5)

mod <- cmdstan_model(model_file, pedantic = TRUE)
fit <- mod$sample(data = stan_data, seed = 1)

ito4303

# Bayesian Workflow Exercise 3.1(b)

library(purrr)

set.seed(123)

# Returns log likelihood
# p: proportion of honest participants
# N: Number of participants
# W: Number of participants who claim prizes

ito4303

# Bayesian Workflow Exercise 3.1

library(ggplot2)
library(purrr)

set.seed(123)

# Log likelihood function
# p: proportion of honest participants
# N: Number of participants

ito4303

# 3D animated map from DEM

library(terra)
library(ggplot2)
library(geometry)
library(ggcube)

# Load DEM data from GeoTIFF
# 国土地理院基盤地図情報数値標高モデル（DEM10B）を使用
geotiff_file <- "data/543654.tif"

ito4303

# Plot DEM data using ggcube

library(terra)
library(ggplot2)
library(geometry)
library(ggcube)

# Load DEM data from GeoTIFF
# 国土地理院基盤地図情報数値標高モデル（DEM5A）を使用
geotiff_file <- "data/543654.tif"

ito4303

# Mexican hat using ggcube
#
# https://tororolab.hatenablog.com/entry/2024/12/16/225348

library(ggplot2)
library(ggcube)

fun <- function(x, y) {
  r <- sqrt(x^2 + y^2)
  1.5 * cos(r) - 0.75 * cos(3 * r)

ito4303

library(ggplot2)
library(palmerpenguins)

point_plot <- function(d, x, y) {
  ggplot(d) +
    geom_point(aes(x = .data[[x]], y = .data[[y]]))
}

(var_names <- colnames(penguins))