benford-prices.r

gistfile1.r

library(ggplot2)
orig <- read.csv("prices.tsv.gz", sep = "\t")
orig <- mutate(orig,
  cents = (100 * Price) %% 100,
  fd = cents %/% 10)

df <- count(orig, "fd")
df <- mutate(df,
  prob = prop.table(freq),
  benford = log10(fd + 1) - log10(fd + 0))

ggplot(df, aes(x = fd, y = prob)) +
  geom_bar(stat = "identity", fill = "blue") +
  geom_line(aes(x = Numeral, y = benford, size = 0.1)) +
  geom_point(aes(x = Numeral, y = benford, color = "red", size = 1)) +  
  theme_bw()
  scale_x_continuous(breaks = seq(1:9))
  
ggsave("benford.png")