There are basically two steps doing things in SAS. one is creating, manipulating data, The DATA step and another is all other functions including statistical, the PROC step.
I will cover both steps and many procedures later but here is quick look for how to use them.
Use DATA STEPS to:
-
Create and import data:
data data1;
input a b c;
cards;
1 2 3
4 5 6
7 8 9
;
run;
The result is the following table
Obs | a | b | c |
---|---|---|---|
1 | 1 | 2 | 3 |
2 | 4 | 5 | 6 |
3 | 7 | 8 | 9 |
data data3;
infile 'D:/sas/data3.dat';
input a b c d;
run;
This will read data3.dat file into data3 which has four variables as given in input statement.
The result is the following tabledata
random1;data random1;
do i = 1 to 10;
x = normal(0);
y = uniform(0);
output;
end;
run;
Obs | i | x | y |
---|---|---|---|
1 | 1 | -1.52857 | 0.41513 |
2 | 2 | 0.08083 | 0.53118 |
3 | 3 | 1.98095 | 0.32572 |
4 | 4 | -0.26286 | 0.38283 |
5 | 5 | -0.01880 | 0.21665 |
6 | 6 | -0.04803 | 0.43361 |
7 | 7 | -1.46001 | 0.46454 |
8 | 8 | 1.19792 | 0.13875 |
9 | 9 | -1.63991 | 0.31687 |
10 | 10 | 1.00099 | 0.07637 |
-
Modify and manipulate data:
data data2; set data1;
d=(a+b)/c;
loga=log10(a);
drop a;
run;
The result is the following table
Obs | b | c | d | loga |
---|---|---|---|---|
1 | 2 | 3 | 1.00000 | 0.00000 |
2 | 5 | 6 | 1.50000 | 0.60206 |
3 | 8 | 9 | 1.66667 | 0.84510 |
The result is the following tabledata
random2; set random1;data random2; set random1;
if x<0 then out1='Neg';
else out1='Pos';
if y<.5 then out2='F';
else out1='P';
run;
Obs | i | x | y | out1 | out2 |
---|---|---|---|---|---|
1 | 1 | -1.52857 | 0.41513 | Neg | F |
2 | 2 | 0.08083 | 0.53118 | Pos | P |
3 | 3 | 1.98095 | 0.32572 | Pos | F |
4 | 4 | -0.26286 | 0.38283 | Neg | F |
5 | 5 | -0.01880 | 0.21665 | Neg | F |
6 | 6 | -0.04803 | 0.43361 | Neg | F |
7 | 7 | -1.46001 | 0.46454 | Neg | F |
8 | 8 | 1.19792 | 0.13875 | Pos | F |
9 | 9 | -1.63991 | 0.31687 | Neg | F |
10 | 10 | 1.00099 | 0.07637 | Pos | F |
Use PROC STEPS to:
- Sort data (PROC SORT)
- Statistical summary (PROC MEANs, PROC UNIVARIATE)
- Graphics (PROC PLOT, PROC GPLOT)
- Many Statistical procedures (PROC GLM, PROC MIXED, PROC PRINCOMP, …)
- Advanced data techniques (PROC SQL)
- Matrix operations (PROC IML)
- ……………and many more
I will discuss many topics later
0 comments:
Post a Comment