Compare and Conquer SAS® Programming

e-Guide

Training Videos:

Five Habits of Highly Effective SAS Programmers

Below is a collection of advantages and disadvantages in similar SAS® programming techniques. Often, a greater understanding of SAS® programming techniques is realized after knowing the details of why one method is better than another method. Hope you also find this blog insightful.

SAS® PROGRAMMING

TECHNIQUE 1

SAS® PROGRAMMING

TECHNIQUE 2

COMPARISON

DATA Step

(Simple 1:1, 1:M Joins, Non-matching/No-By Variable Joins or Similar structure/Horizontal Join)

PROC SQL

(Adding new variables, Fuzzy or Complex Joins)

Paper, Example

Lookup Techniques

Correct DATA Step Merge

Incorrect DATA Step Merge

Paper 1,
2

Do-Loops (Top Evaluation)

Do-Loops (Bottom Evaluation)

INFILE'S MISSOVER

INFILE'S TRUNCOVER

Reference, Paper, 2

PROC MEANS

PROC SQL

Paper , 2

PROC MEANS

(displays results by default)

PROC SUMMARY

(almost always used with OUTPUT OUT= to save results to dataset)

DATA _NULL_

Paper, 2

PUT()

INPUT()

Reference

WHERE

Summary Sheet, SAS Tip

Training Video

data allab; merge a (in=a) b (in=b); by id; if a then a=1; if b then b=1; run;

data allab alla anotb aandb bnota allb; merge a (in=a) b (in=b); by id; if a or b then output allab; if a then output alla;

if a not b then output anotb; if a and b then output aandb; if b not a then output bnota; if b then output allb; run;

Topic

Discussion

INFILE'S MISSOVER vs. TRUNCOVER

SAS Paper

TRUNCOVER and MISSOVER are useful alternatives to FLOWOVER. In both, SAS does not move to
the next line of input data should the INPUT statement request more data than exists on the current line. Instead, it stops, and sets the remainder of the requested variables on the line to missing. The difference between the two options is how it handles the situation of a line ending in the middle of populating a variable. MISSOVER sets that variable to missing along with the others, while TRUNCOVER “settles for what it can get” and populates the variable with the last X bytes of data that were read in from the line. There is another alternative for those who prefer more extreme preventative measures – STOPOVER.
This option basically ceases execution of the DATA step upon occurrence of the “end of data on a line”
condition, and trips the error flag.

PUT() vs. INPUT()

INPUT is typically used to convert a character value into a numeric value, but it can also be used to convert a character value into a different character value, while PUT is typically used to convert a numeric value into a character one.