if (!require(devtools)) {
	install.packages("devtools")
	library(devtools)
}
devtools::install_github('paasim/glmproj', build_vignettes = TRUE)

rm(list=ls())
library(glmproj)
library(rstanarm)
options(mc.cores = parallel::detectCores())
set.seed(1)

# Gaussian and Binomial examples from the glmnet-package
data('df_gaussian', package = 'glmproj')
#data('df_binom', package = 'glmproj')

# fit the full model with a sparsifying prior
fit <- stan_glm(y ~ x, family = gaussian(), data = df_gaussian,
                prior = hs(df = 1, global_scale=0.03), iter = 500, seed = 1)
#fit <- stan_glm(y ~ x, family = binomial(), data = df_binom
#                prior = hs(df = 1, global_scale=0.03), iter = 500, seed = 1)


# perform the variable selection
fit_v <- varsel(fit)

# print the results
varsel_statistics(fit_v)

# project the parameters for model sizes nv = 3,5 variables 
fit_p <- project(fit_v, nv = c(3, 5))
proj_coef(fit_p)

# perform cross-validation for the variable selection
fit_cv <- cv_varsel(fit, cv_method='LOO')

# plot the results
varsel_plot(fit_cv)