Quiz 1 (10 minutes, 3 points):

Read-in data for problem 1.16 from
http://www.stat.washington.edu/marzban/390/1_16_dat.txt .
a) Write code for plotting the histogram of the data with 15 breaks, and
b) for the histogram of the logarithm (base 2) of the data.
Make sure your code works! Before you log out, cut and paste your code
into an email to your TA. The whole code should be no longer than 10 lines long.

Soln:

dat = read.table("http://www.stat.washington.edu/marzban/390/1_16_dat.txt", header=F)
x = dat[,1]
hist(x,breaks=15)         # Skewed to the left.
y = log2(x)
hist(y,breaks=15) 
or hist(log2(x),breaks=15)

# Note that log-transform makes histograms more normal. This is important,
# because many of the techniques we will develop in class assume that the
# distribution of the variable is normal. This log-transform trick is just
# one example of transformations one can do to a variable to make its
# distribution normal.