ST 610N -- Fall 2006
       Homework 1
       Due Tuesday, 12 September 2006
       

       EXERCISE 

       Use the following data that's also in the file 
          infmnc2k.dat 
       in the class directory.  Write a SAS program to read
       in the county name and lots of information relevant to
       understanding patterns of infant mortality in North Carolina:
         county name                       cols  1-12
         white infant deaths                    13-16
         nonwhite infant deaths                 17-21
         total infant deaths                    22-27
         white births                           28-33
         nonwhite births                        34-39
         total births                           40-45
         high school graduation rate (%)        46-51
         unemployment rate (%)                  52-56
         per capita personal income ($)         57-62
         median family income ($)               63-68

 ALAMANCE     13   17    30  1510   392  1902  67.9  2.8 26679 53100 
 BUNCOMBE     17    6    23  2318   279  2597  74.5  2.6 27393 46800

 WAYNE        11    9    20  1104   659  1763  71.2  4.0 20050 43200
 WILSON        3    8    11   609   471  1080  62.2  7.0 24550 47000

       a) Create a SAS dataset with the county name (COUNTY) and the 
       three economic variables (UNEMP, PCPI, MFAMINC).  Include three 
       new variables:
       BIG    indicates that a county has a large number of births (>5000)
       INFMOR gives the infant mortality rate (deaths/1000 births)
       FNW    gives the fraction of births that are nonwhite (nwb/births) 
 
       b) Include code to do some checks on the data -- the two death 
       counts should add up, as well as the two counts for births.  

       c) Print the data using PROC PRINT and include an appropriate
       title.  The syntax of a TITLE statement is TITLE ' -text-  ' ;
       The quote is the right single quote mark, on many keyboards
       it's on the same key as the double quote ",  and -text- stands
       for any text you want to put in.  Print just the following
       variables (and in this order): county name, infant mortality
       rate, BIG, and the three economic variables (UNEMP, PCPI, MFAMINC).
       (If you created a variable to do the checks, you can include
       this in your list -- but try to do the checks without printing
       anything out.)

       d) Use the code
         PROC PLOT DATA=whatever ; 
           PLOT INFMORT*PCPI ; 
       to plot the infant mortality versus pcpi.  Notice 
       that you cannot determine the county.  Now try 
         PROC PLOT DATA=whatever; 
           PLOT INFMORT*PCPI=COUNTY; 

       (You may want to adjust the page size for these plots --
         OPTIONS PS=40 ;
       uses 40 lines for a SAS output page.)

       FOR ALL EXERCISES: Hand in
        (1)  Your program
        (2)  Your output