.rs.restartR()
library(cmdstanr)
library(posterior)
stan_dir <- "/Users/mwatson/Dropbox/TVExtreme/Disasters/Stan/CmdStanR/"
stan_file <- stan_file.path(stan_dir, "Static_Model.stan")
stan_file <- file.path(stan_dir, "Static_Model.stan")
mod <- cmdstan_model(stan_file)
stan_dir <- "/Users/mwatson/Dropbox/TVExtreme/Disasters/Stan/CmdStanR"
stan_file <- file.path(stan_dir, "Static_Model.stan")
mod <- cmdstan_model(stan_file)
mod <- cmdstan_model(stan_file)
mod
data_file <- file.path(stan_dir, "EventsMonthly_1980_2023.data.json")
init_file <- file.path(stan_dir, "4RW_Model.init.json")
mod$sample(
data file = data_file,
mod$sample(
data = data_file,
num_warmup = 100,
num_samples = 1000,
init=init_file,
seed = 76161431
)
mod$sample(
data = data_file,
iter_warmup = 100,
iter_sampling = 1000,
init=init_file,
seed = 76161431
)
fit$summary()
fit <- mod$sample(
data = data_file,
iter_warmup = 100,
iter_sampling = 1000,
init=init_file,
seed = 76161431
)
fit$summary()
fit$sampler_diagnostics()
fit$diagnostic_summary()
draws_list()
draws_matrix()
draws_matrix("m")
fit <- mod$sample(
data = data_file,
iter_warmup = 100,
iter_sampling = 1000,
init=init_file,
cores = 8,
seed = 76161431
)
fit <- mod$sample(
data = data_file,
iter_warmup = 1000,
iter_sampling = 10000,
init=init_file,
chains = 4,
parallel_chains = 4,
seed = 76161431
)
fit <- mod$sample(
data = data_file,
iter_warmup = 1000,
iter_sampling = 10000,
init=init_file,
refresh = 1000,
chains = 4,
parallel_chains = 4,
seed = 76161431
)
draws_arr <- fit$draws()
tmp <- subset_draws(draws_arr, c("xi","psi","lambda"))
tmp_m <- merge_chains(tmp)
tmp$variable
ls tmp
print(tmp$variable)
tmp <- subset_draws(draws_arr, c("xi"))
xi_draws <- merge_chains(tmp)
tmp <- subset_draws(draws_arr, c("psi"))
psi_draws <- merge_chains(tmp)
tmp <- subset_draws(draws_arr, c("lambda"))
lambda_draws <- merge_chains(tmp)
tmp1 <- subset_draws(draws_arr, c("xi"))
xi_draws <- merge_chains(tmp1)
tmp2 <- subset_draws(draws_arr, c("psi"))
psi_draws <- merge_chains(tmp2)
tmp3 <- subset_draws(draws_arr, c("lambda"))
lambda_draws <- merge_chains(tmp3)
tmp <- subset_draws(draws_arr, c("xi"))
xi_draws <- merge_chains(tmp)
tmp <- subset_draws(draws_arr, c("psi"))
psi_draws <- merge_chains(tmp)
tmp <- subset_draws(draws_arr, c("lambda"))
lambda_draws <- merge_chains(tmp)
setwd("~/Dropbox/Shared_Folders/SpatialUR/ReplicationPackage_MS21654_MuellerWatson_r1/Data/AdjacencyMatrix_cz1990")
library(tidyverse)
install.packages("tidyverse")
install.packages("sf")
install.packages("furrr")
library(tidyverse)
library(sf)
library(furrr)
# Raw Data Downloaded from "https://healthinequality.org/data/"
cz1990 <- st_read("cz1990_shapefile/cz1990.shp")
crosswalk <- read_csv("cty_cz_st_crosswalk.csv")
# Exclude Alaska and Hawaii
cz_AH <- crosswalk %>% filter(statename %in% c("Alaska","Hawaii")) %>%
select(cz) %>% distinct()
cz_dta <- cz1990 %>% filter(!cz %in% cz_AH$cz) %>% arrange(cz) %>% mutate(cz = as.character(cz))
# Map Illustration
ggplot() +
geom_sf(data = cz_dta, fill = "lightblue", color = "black", size = 0.2)+
theme_classic() +
labs(title = "Shapefile Plot")
# Find Adjacency
plan(multisession)  # Use multiple cores for parallel processing
cz_dta <- cz_dta %>%
st_transform(crs = 4326) %>%  # Ensure a common CRS for the spatial operations
mutate(
neighbor_ids = future_map(st_touches(., .), function(neighbors) {
if (length(neighbors) > 0) {
cz_dta$cz[unlist(neighbors)]
} else {
NA
}
})
)
# Create Adjacency Matrix
all_nodes <- unique(cz_dta$cz)
adj_matrix <- matrix(0, nrow = length(all_nodes), ncol = length(all_nodes))
rownames(adj_matrix) <- colnames(adj_matrix) <- all_nodes
for (i in 1:nrow(cz_dta)) {
location <- cz_dta$cz[i]
neighbors <- cz_dta$neighbor_ids[[i]]
adj_matrix[location, neighbors] <- 1
adj_matrix[neighbors, location] <- 1
}
write.csv(adj_matrix, file = "CZadj_matrix.csv")
