HW1: [PG] Chapter 2, pages 30 - 34.
Exercises 7, 8, 10, 11, 13, 14, 15

Exercise 7
a. discrete
b. continuous
c. continuous
d. discrete

Exercise 8
c. The histogram is unimodal, and is fairly symmetric. There is a box
to the right, representing the health expenditure in the US, that 
is disjoint with the main part of the histogram, which shows that 
the health expenditure in the US is much larger than the others. 

Exercise 13
a. Because the total number of smokers is not equal to the total number
of nonsmokers, it is not fair to compare the distributions of absolute
frequencies for these two groups. 

b. 

c. 

d. The distribution of smokers is fiarly uniform across cotinine levels. 
The relative frequency is smallest in the first interval (0-13ng/ml). It 
then increases, and remains consistent (hovering around 10%) across 
subsequent intervals up to the last (300+ ng/ml), where the relative frequency
in each of the other intervals is very small.

e. Yes, it is possible that some of the subjects are misclassified. In 
particular, there are a number of self-reported ``nonsmokers'' with
very high cotinine levels. 

Exercise 14
a. The workers have lower blood lead levels in 1987.
b. The cumulative relative frequencies for each group of workers 
are displayed below:

        Blood Lead    1979     1987
         (mug/dl)     (%)      (%)
        ----------------------------
          < 20        11.5     37.8
         20-29        23.6     52.5
         30-39        37.5     65.5
         40-49        52.9     80.9
         50-59        69.4     91.4
         60-69        82.2     98.2
         70-79        90.6     99.6
          =>80       100.0    100.0
        ---------------------------
c. The distribution of blood lead levels is stochatically larger
for the group of workers in 1979. 

Exercise 15

b. Based on this two-year period, there appears to be more births
in the spring and summer months and fewer in fall and winter.