* Array problems ;
* Problem 1 ;
* Often you are given data with missing values that should be treated
as zeros, or data coded with zeros that should be treated as
missing. An example (from the SAS documentation) is some data on
the number of different types of books (SAS Documentation!) sold
in several different locations:
Location Reference Usage Introduction
A 45 63 113
B . 75 150
C 62 . 98
;
* Input these data, use an ARRAY to recode, and calculate the proportion
of "Usage" books sold (the marginal proportion). Hint: use
PROC FREQ and weights to get the proportion. Why are the zeros
necessary? ;
* Problem 2 ;
* Arrays provide a quick way to define a group of similarly named
variables. Here are data of noon temperatures in two cities -
each line gives five temperatures (Monday through Friday) from
one city, and the two cities alternate lines (each pair of lines
is the data from one week). Use an array to input the data,
and reshape it so that you can calulate the mean and standard
deviation for each city. Do a t-test (or the analogous F-test)
to determine a mean difference and confidence interval for the
difference.
City1 89.5 65.4 75.3 77.7 89.3
City2 73.7 87.3 89.9 98.2 35.6
City1 75.8 82.1 98.2 93.5 67.7
City2 101.3 86.5 59.2 35.6 75.7
;