This article explains practical application of SQL queries with examples.
Example 1 : Creating trend variables
Output :
1. ANY operator selects values that pass the comparison test with any of the values that are returned by the sub-query.
Example 1 : Creating trend variables
 |
| Sample Dataset |
data example1;Task : Calculate total revenue and total balance accumulated in the first 3 months and how much the data spreads in the first 3 months time period.
input ID Months Revenue Balance;
cards;
1011390
10123368
10132251
1014390
10153365
10162254
102110018
10225862
10239597
102410018
10255865
10269592
;
proc sql noprint;
create table output1 as
select ID,
sum(case when 1 <= Months <= 3 then Revenue else . end) as Rev_1_3,
sum(case when 1 <= Months <= 3 then Balance else . end) as Bal_1_3,
var(case when 1 <= Months <= 3 then Revenue else . end) as Var_Rev_1_3,
var(case when 1 <= Months <= 3 then Balance else . end) as Var_Bal_1_3
from example1
group by ID;
quit;
Output :
 |
| Output |
Example 2 : Extract First and Last Observation within a Group
PROC SQL : Alternative to First. Statement
proc sql;
create table output2 (drop=n) as
select *, monotonic() as n
from example1
group by ID
having min(n) = n;
quit;
PROC SQL : Alternative to Last. Statement
proc sql;
create table output2 (drop=n) as
select *, monotonic() as n
from example1
group by ID
having max(n) = n;
quit;
Example 3 : Self Join
Suppose you have data for employees. It comprises of employees' name, ID and manager ID. You need to find out manager name. |
| Proc SQL : Self Join |
data example2;
input Name $ ID ManagerID;
cards;
Smith123456
Robert456 .
William222456
Daniel777222
Cook383222
;
run;
SQL Query : Self Join
proc sql;
create table want as
select a.*, b.Name as Manager
from example2 as a left join example2 as b
on a.managerid = b.id;
quit;
Example 4 : Capping Extreme Values
Suppose you need to cap values of a column.
data have1;
input x y z;
cards;
101 1 10
102 2 20
103 3 45
104 1 23
105 2 42
106 3 46
107 1 61
109 2 22
110 3 28
111 1 30
112 2 32
113 3 39
;
run;
proc sql noprint;
create table output2 (drop=z rename= (z1=z)) as
select *, case when z >= 40 then 40 else z
end as z1 from have1;
quit;
First cap values of column z and then sum all the z values by column y.
proc sql;
select sum(z1) as OutCol
from (select *, case when z >= 40 then 40 else z
end as z1 from have1)
group by y;
quit;
Example 5 : Select All Excluding 1 Variable
proc sql;Example 6 : Sub Query - ANY and ALL Operators
create table want (drop=x) as
select a.*,b.*
from dat1 a, dat2 (rename=(id=x)) b
where a.id = b.x;
quit;
2. ALL operator selects values that pass the comparison test with all of the values that are returned by the sub-query.
data jansale;
input sale id;
cards;
1001
105 2
1083
1104
;
run;
data febsale;
input sale id;
cards;
1201
105 2
1183
1174
;
run;
proc sql;
select *
from jansale
where sale < all (select sale from febsale);
quit;
proc sql;
select *
from jansale
where sale < any (select sale from febsale);
quit;