ericnovik

par (mar=c(3,3,2,1), mgp=c(2,.7,0), tck=-.012, las=1)
with(iris, plot(Sepal.Length, Sepal.Width, 
                col=as.numeric(Species)+1, pch=20))
lbs <- levels(iris$Species)
legend('topright', legend=lbs, 
       col=2:4, cex=0.7, pch=20, box.lwd=0.5, pt.cex=0.6)

ericnovik

order <- 10
x <- seq(0, 2*pi, 0.01)
y <- cos(x)

y_poly <- poly(y, order, raw = TRUE)
plot(y_poly[, 1], col = 1, type = "l", xlab = "", ylab = "", axes = FALSE)

for (i in 2:order) {
  lines(y_poly[, i], col = i)
}

ericnovik

require('manipulate')

manipulate( 
{
  x <- seq(-10, 10, length = 1000)
  d <- dcauchy(x, scale = scale, location = location, log = Log)
  plot(d ~ x, type = "l")
},

scale = slider(1, 10, step = 1),

ericnovik

var_imp <- function(rpart_fit) {
  var_imp <- sort(rpart_fit$variable.importance)
  in_model <- levels(rpart_fit$frame$var)[-1]
  
  return(list(var_imp = var_imp, in_model = in_model))
}

var_imp_plot <- function(rpart_fit) {
  vars <- var_imp(rpart_fit)
  groups <- factor(x = rep("less so", length(vars$var_imp)), 

ericnovik

Verifying that +ericnovik is my openname (Bitcoin username). https://onename.io/ericnovik

ericnovik

# Example 2.2.7 (A girl born in winter). Page 46.
# A family has two children. Find the probability that both children are girls, 
# given that at least one of the two is a girl who was born in winter. Assume 
# that the four seasons are equally likely and that gender is independent of season.
# Blitzstein, Joseph K.; Hwang, Jessica (2014-07-24). 
# Introduction to Probability (Chapman & Hall/CRC Texts in Statistical Science) 

library(stringr)
n <- 1000000

ericnovik

read_cmd_stan_output <- function (dir, file_prefix, n_chains = 4) {
  csvfiles <- character(n_chains)
  for (i in 1:n_chains)
    csvfiles[i] <- paste0(dir, file_prefix, i, ".csv")
  stan_fit <- rstan::read_stan_csv(csvfiles)
  return(stan_fit)
}

# Assume /my/base/dir/ contains stan_output_model_1.csv, 
# stan_output_model__2.csv, stan_output_model_3.csv, 

ericnovik

data <- list(N = 5, y = c(0, 1, 1, 0, 1))

# log probability function
# Note: the below operations are not arithmetically 
# stable. Use for demo purposed only.
lp <- function(theta, d) {
  lp <- 0
  for (i in 1:d$N) {
    lp <- lp + log(theta) * d$y[i] + 
      log(1 - theta) * (1 - d$y[i])

ericnovik

library(ggplot2)
entropy <- function(p) -sum(p * log(p))
n <- 1e3
p <- runif(n)
q <- 1 - p
z <- matrix(c(p, q), ncol = 2)
ent <- apply(z, 1, entropy)
qplot(q, ent, geom = "line") + 
  ylab("Entropy") + xlab("Probability") +
  ggtitle("Entropy For Two Events")

ericnovik

library(ggplot2)
theme_set(theme_minimal())

# Visualizing entropy
entropy <- function(p) -sum(p * log(p))
n <- 1e3
p <- runif(n)
q <- 1 - p
z <- matrix(c(p, q), ncol = 2)
ent <- apply(z, 1, entropy)