Course 678: Analysis of Multivariable Data.  June 1999

         Homework to be handed in by Wednesday June 9
         
*****************************************************************


1.   KKMN Ch 9,  Question 7 (p155-6 and 126-7).


2.   KKMN Ch 9,  Question 11 (p.158 and p132-3)


3    See the documentation on the "bodyfat data (Small)"
     on the class web page
          
     Download the dataset, say to your "a" drive
     
     Paste the following SAS  program into the Program Editor
     
        OPTIONS LS=75 PS=60;
        DATA sasuser.bodyfat0;
        INFILE 'a:\bodyfat0.dat' ;
        INPUT  Triceps Thigh Midarm BodyFat  ;
        RUN;
     
     and click on the "run" icon to produce and save a sas file
     called "bodyfat0".
     
     a. Reproduce the calculations in the documentation.

     b. Which single-variable model has the highest r-square?
     
        Does the addition of the better of the other 2 remaining
        variables add significantly to it?
        
        As a layperson, with no particular expertise in this subject,
        can you think of other easy-to-obtain variables that might
        help in the predictions (besides the three you are given, and 
        not counting powers or other functions of the 3)?
        
        You might want to glance at the documentation for the other 
        (larger) dataset on body fat. -- No calculations required at
        this stage!