SAS Problems

In this blog post I will be presenting problems of  SAS from a textbook written by Ron Cody.
The flow will be as such

• Problem statement
• Data description
• Code
• Result
• Learning - what I have learnt from this problem

I am working on a permanent dataset with library name A15033

The link is below

libname A15033 "/folders/myfolders";

To access data sets click here

PROBLEM 1 :

Starting with the Blood data set, create a new, temporary SAS data set containing all the variables in Blood plus a new variable called CholGroup. Define this new variable as 

CholGroup 

Chol  low             low -110

          medium     111 - 140 

          high           141 - high

Use a SELECT statement to do this.

DATA DESCRIPTION;

We have a dataset named blood.txt. It contains id, name, bloodgroup, agegroup, wbc,rbc,chol.


CODE

libname A15033 "/folders/myfolders";
data A15033.blood;
infile "/folders/myfolders/blood.txt";
input id $ gender $ bloodgroup $ agegroup $ wbc rbc Chol;
 select;
 when (missing(Chol)) CholGroup = ' ';
 when (Chol le 110) CholGroup = 'Low';
 when (Chol le 140) CholGroup = 'Medium';
 otherwise CholGroup = 'High';
end; run;

proc print data=A15033.blood (obs=10);
run;

RESULT

Click on the image to enlarge

LEARNING

In this program I learned about how to assign group names based on their characteristics and how to treat missing values. 

PROBLEM 2;

 Using the Bicycles data set, list all the observations for Road Bikes that cost more than $2,500 or Hybrids that cost more than $660. The variable Model contains the type of bike and UnitCost contains the cost


DATA DESCRIPTION

This program is executed using datalines, So the data will be  in the program.


CODE

data bicycles;
input @1 country $15.
      @16 model   $15.
      @31 purpose $11.
      @42 quantity $6.
      @48 price dollar11.;
   
   
datalines;
USA             Road Bike      Trek       5000  $2200
USA             Road Bike      Cannondale 2000  $2100
USA             Mountain Bike  Trek       6000  $1200
USA             Mountain Bike  Cannondale 4000  $2700
USA             Hybrid         Trek       4500  $650
France          Road Bike      Trek       3400  $2500
France          Road Bike      Cannondale 900   $3700
France          Mountain Bike  Trek       5600  $1300
France          Mountain Bike  Cannondale 800   $1899
France          Hybrid         Trek       1100  $540
United Kingdom  Road Bike      Trek       2444  $2100
United Kingdom  Road Bike      Cannondale 1200  $2123
United Kingdom  Hybrid         Trek       800   $490
United Kingdom  Hybrid         Cannondale 500   $880
United Kingdom  Mountain Bike  Trek       1211  $1121
Italy           Hybrid         Trek       700   $690
Italy           Road Bike      Trek       4500  $2890
Italy           Mountain Bike  Trek       3400  $1877
;
run;

title "Selected Observations from BICYCLES";
proc print data=bicycles noobs;  
where model eq "Road Bike" and price gt 2500 or        
model eq "Hybrid" and price gt 660;  
run;

RESULT

LEARNING

In this program I have learned how to subset require information from data using "where" option



PROBLEM 3;

 Run the program here to create a temporary SAS data set called Vitals:
   data vitals;    
   input ID    : #3.             Age                   Pulse                 SBP             DBP;    
   label SBP = "Systolic Blood Pressure"          
            DBP = "Diastolic Blood Pressure";  
datalines;  
001 23 68 120 80  
002 55 72 188 96  
003 78 82 200 100  
004 18 58 110 70  
005 43 52 120 82  
006 37 74 150 98  
007  . 82 140 100    ;
Using this data set, create a new data set (NewVitals) with the following new variables:
    For subjects less than 50 years of age:
        If Pulse is less than 70, set PulseGroup equal to Low;    
        otherwise, set PulseGroup equal to High.      
        If SBP is less than 130, set SBPGroup equal to Low;      
        otherwise, set SBPGroup equal to High.
    For subjects greater than or equal to 50 years of age:
       If Pulse is less than 74, set PulseGroup equal to Low;      
       otherwise, set PulseGroup equal to High.      
       If SBP is less than 140, set SBPGroup equal to Low;      
       otherwise, set SBPGroup equal to High.
 assume there are no missing values for Pulse or SBP.


DATA DESCRIPTION

The data is given in the problem

CODE

data vitals;  
input ID    : $3. Age Pulse SBP DBP;  
label SBP = "Systolic Blood Pressure"        
      DBP = "Diastolic Blood Pressure";
datalines;
001 23 68 120 80
002 55 72 188 96
003 78 82 200 100
004 18 58 110 70
005 43 52 120 82
006 37 74 150 98
007  . 82 140 100
;
run;

data newvitals;  
set vitals;  
if Age lt 50 and not missing(Age) then do;    
   if Pulse lt 70 then PulseGroup = 'Low ';    
   else PulseGroup = 'High';    
   if SBP lt 140 then SBPGroup = 'Low ';    
   else SBPGroup = 'High';  
end;  
else if Age ge 50 then do;    
   if Pulse lt 74 then PulseGroup = 'Low';    
   else PulseGroup = 'High';    
   if SBP lt 140 then SBPGroup = 'Low';    
   else SBPGroup = 'High';  
end;
run;

title "Listing of NEWVITALS";
proc print data=newvitals noobs;
run;

 

RESULT

LEARNING

I have learned how to use nested if else statements along with do statement

PROBLEM 4:

 Modify the program here so that each observation contains a subject number (Subj), starting with 1: 

   data test;       

   input Score1-Score3;       

   /* add your line(s) here */    

   datalines;    

   90 88 92    

   75 76 88    

   88 82 91    

   72 68 70

   ;

   run;   

DATA DESCRIPTION

The date is given is in the program itself

CODE

data test;
 input Score1-Score3;
 Subj + 1;
 datalines;
 90 88 92
 75 76 88
 88 82 91
 72 68 70
 ;
 run;

 title "Listing of TEST";
 proc print data=test noobs;
 run;

RESULT

LEARNING

I learnt how to sum and list observations



PROBLEM 5;

 You have the following seven values for temperatures for each day of the week, starting with  Monday: 70, 72, 74, 76, 77, 78, and 85. Create a temporary SAS data    set (Temperatures) with a  variable (Day) equal to Mon, Tue, Wed, Thu, Fri, Sat, and    Sun and a variable called Temp equal to  the listed temperature values. Use a DO    loop to create the Day variable.



DATA DESCRIPTION

The data is in the program itself


CODE

data temperatures;
do Day = 'Mon','Tues','Wed','Thu','Fri','Sat','Sun';
 input Temp @;
 output;
 end;
 datalines;
 70 72 74 76 77 78 85
 ;
run;

 title "Listing of TEMPERATURES";
 proc print data=temperatures noobs;
 run;

RESULT

LEARNING

I learnt how to assign days in chronological order using do loop

PROBLEM  6:
 You invest $1,000 a year at 4.25% interest, compounded quarterly. How many   years will it take to reach $30,000? Do not use compound interest formulas. Rather,   use “brute force” methods with DO WHILE or DO UNTIL statements to solve this  problem.

DATA DESCRIPTION

The data is in the problem statement

CODE

data money;  
do Year = 1 to 999 until (Amount ge 30000);    
  Amount + 1000;    
  do Quarter = 1 to 4;        
    Amount + Amount*(.0425/4);    
    output;    
  end;  
end;  
format Amount dollar10.;
run;

title "Listing of MONEY";
proc print data=money (obs=10);
run;

RESULT

To view as a image I have made observations to 10 but the result will be a bit longer than that

LEARNING

I have learned how to use DO WHILE and DO UNTIL statements to calculate compound interest

PROBLEM 7;

 You have several lines of data, consisting of a subject number and two dates (date of   birth and visit date). The subject starts in column 1 (and is 3 bytes long), the date of birth starts in column 4 and is in the form mm/dd/yyyy, and the visit date starts in  column 14 and is in the form nnmmmyyyy (see sample lines below). Read the  following lines of data to create a temporary SAS data set called Dates. Format both  dates using the DATE9. format. Include the subject’s age at the time of the visit in  this data set.
0011021195011Nov2006  
0020102195525May2005  
0031225200525Dec2006


DATA DESCRIPTION

The data is given in the program itself

CODE

data dates;  
input @1  Subj  $3.        
      @4  DOB   mmddyy10.        
      @14 Visit date9.;  
Age = yrdif(DOB,Visit,'Actual');  
format DOB Visit date9.;
datalines;
00110/21/195011Nov2006
00201/02/195525May2005
00312/25/200525Dec2006
;


title "Listing of DATES";
proc print data=dates noobs;

run;

RESULT

LEARNING
I learnt how to calculate the age difference between two dates







PROBLEM 8:
 Using the SAS data set Blood, create two temporary SAS data sets called Subset_A and Subset_B.  Include in both of these data sets a variable called Combined equal to .001 times WBC plus RBC.  Subset_A should consist of observations from Blood where Gender is equal to Female and  BloodType is equal to AB. Subset_B should consist of all observations from Blood where Gender is  equal to Female, BloodType is equal to AB, and Combined is greater than or equal to 14.



DATA DESCRIPTION
The data set blood contains variables such as name, id , bloodgroup, agegroup, wbc, rbc, chol etc.




CODE

data subset_a;  
set mozart.blood;  
where Gender eq 'Female' and bloodgroup='AB';  
Combined = .001*WBC + RBC; run;


title "Listing of SUBSET_A";
proc print data=subset_a noobs;
run;


data subset_b;  
set mozart.blood;  
Combined = .001*WBC + RBC;  
if Gender eq 'Female' and bloodgroup='AB' and Combined ge 14;
run;


title "Listing of SUBSET_B";
proc print data=subset_b noobs;

run;

 

RESULT

LEARNING

In this program I learnt how to subset from a data set using where statement



PROBLEM 9;


Using the SAS data set Health, compute the body mass index (BMI) defined as the weight in kilograms divided by the height (in meters) squared. Create four other variables based on BMI: 1) BMIRound is the BMI rounded to the nearest integer, 2) BMITenth is the BMI rounded to the nearest tenth, 3) BMIGroup is the BMI rounded to the nearest 5, and 4) BMITrunc is the BMI with a fractional amount truncated. Conversion factors you will need are: 1 Kg equals 2.2 Lbs and 1 inch = .0254 meters.


DATA DESCRIPTION

The health data set consists of id, date of birth ,weight, height


CODE

libname A15033 "/folders/myfolders";
data A15033.health;
infile "/folders/myfolders/health.txt";
input @1 id $3.
      @4 dob mmddyy10.
      weight 14-16 height 17-18;
BMI = (Weight/2.2) / (Height*.0254)**2;  
BMIRound = round(BMI);  
BMIRound_tenth = round(BMI,.1);  
BMIGroup = round(BMI,5);  
BMITrunc = int(BMI);    
run;

title "listing of health";
proc print;
format dob date9.;

run;

RESULT

LEARNING

In this program I learned how to round and truncate integers

PROBLEM 10;

List the first 10 observations in data set Blood. Include only the variables Subject, WBC (white blood cell), RBC (red blood cell), and Chol. Label the last three variables “White Blood Cells,” “Red Blood Cells,” and “Cholesterol,” respectively. Omit the Obs column, and place Subject in the first column. Be sure the column headings are the variable labels, not the variable names. 

DATA DESCRIPTION

The data set blood contains variables such as name, id , bloodgroup, agegroup, wbc, rbc, chol etc.

CODE

data labelling;
set mozart.blood;
run;

title "First 10 Observations in BLOOD";
proc print data=mozart.blood(obs=10);
var   id wbc rbc Chol;  
label wbc = 'White Blood Cells'        
      rbc = 'Red Blood Cells'        
      Chol = 'Cholesterol';

run;

RESULT

LEARNING

In this program I learned how to label variable and column names