Stat $6605$: SAS Programming

Final Exam

Due Date: $04/30/2024$

Please Note:

$1.$ Write your name in your program using a comment statement.

$2.$ Assign a title with the problem number for each problem,

$3.$ Your output for each problem should contain only what is asked.

$4.$ Please submit your programs and outputs.

Problems:

$1.$ Given the set of numbers: $11,23,17,39,50,$ using the array to find averages of any two numbers from a set of numbers. The two numbers can be the same.

$.$

$2.$ Given the data set, using array to change all zeros to missing.

data dd;

input x name $ address $ mm cc di $ kk;

cards;

$8$ aa klkk $0456012$

$0$ bb lkj $12920$

$0$ ddd and $0051$

;

Run;

$3.$ What, if anything, is wrong with the following programs? An ellipsis $(...)$ indicates proportions of a statement that are not germane to the exercise.

a$.$ data p$1$; set mast;

if sector $=$a$4$ then output;

data p$2$; set me;

if sector $=29$ then output;

data sects;

merge p$1$ p$2$;

by sector;

b$.$ data one; infile $...$ ;

length type $2$;

input lab grp m$1-$m$50$;

run;

data two; infile $...$ ;

input lab grp type mx$1-$mx$10$;

run;

data merge;

merge two one;

by lab grp;

run;

$4.$ What, if anything, is wrong with the following programs?

a$.$ data new; update one two three; by id;

b$.$ data new; update transact master; by id;

$5.$ Refer to the data

data demog;

input id contry $ $5-17$ pop $19-22$ pcturb $24-25$ energy $27-29$ cult $31-32$

pasture $34-35$ forest $37-38$;

cards;

$101$ China $10623219113014$

$107$ Japan $1227611113268$

$113$ Philippines $6240938440$

$102$ India $80025751421$

$112$ Vietnam $6219420140$

$109$ Bangladesh $107.....$

$105$ Indonesia $175.....$

$110$ Paskistan $1052862564$

$304$ United States $24374280202628$

$311$ Mexico $827050133823$

$406$ Brazil $141711991966$

$508$ Nigeria $109287342316$

$603$ USSR $28465176101742$

$614$ West Germany $6185163301929$

$615$ Italy $577290411621$

;

run;

and format:

proc format;

value cont $100-199=$'Asia' $200-299=$'Australasia'

$300-399=\text{'}$N$.$ America' $400-499=\text{'}$S$.$ America'

$500-599=$'Africa' $600-699=$'Europe';

Run;

a$.$ Write chart procedure statements to produce a vertical bar chart of average energy consumption by country grouped into continent. Use the cont format to recode country number into continent name. What would the chart look like if the format were not used?

b$.$ How would you rewrite the cont format to arrange the bars in alphabetical order? You could accomplish this by using a chart statement.

$6.$ What, if anything, is wrong with the following assignment statements?

a$.$ xt$=02/$feb$/80$d;

b$.$ baseline$=$d$22$apr$90$;

c$.$ st$=12$mar$58$;

$7.$ Variable zipcode is read with format$$10.$ It reads zip codes in the form $07417$ or $07417-1280.$ Create variable$\_9$digit. It equals $1$ if zipcode has a hyphen separating the fifth and seven digits. Otherwise it equals $0.$ Write the statements three ways, using the length, index, and substr functions.

$8.$ Variable name fullname $($length $$30)$ has a respondent$$s first, middle, and last name. Create var shortname $($length $$20),$ which has only the first and last name, separated by a single blank.

$9.$ Count the number of times the string $?$ occurs in variable line. Choose any method that works reliably. Regardless of method, assume that a line won$$t consist entirely of $?$s$.$

$10.$ A SAS dataset has the following values for its $35$ numeric variables:

$99$ Do not know

$98$ Refused to answer

$97$ Skipped question

Write the data step statement to convert these values to the standard sas missing value $(.)$