Beyond PROC FREQ: Generating High-Performance Frequency Tables with SAS Viya Actions

Difficulty Level

Beginner

Published on : 19/08/2021

Expert Advice

Simon
Expert SAS et fondateur.

Using a computed 0/1 variable as a Weight (as shown in your code with Count) is a powerful alternative to using a WHERE clause. A WHERE clause physically removes observations from the analysis entirely, whereas a weight of 0 simply excludes that specific observation from the frequency count while keeping the row available for other simultaneous calculations or actions that might need it.

Use weighting to filter your analysis without physically removing data.

The example proceeds in two steps. First, a DATA step is used to create a new table `mycaslib.qualifyapps` based on an existing table `mycaslib.creditqualify`. A new variable `Count` is added, taking the value 0 if 'Credit_Qualification' is 'N/A' and 1 otherwise. Second, the `freqTab.freqTab` action of PROC CAS is employed to generate a frequency table for the 'Credit_Qualification' variable, using the 'Count' variable as a weight. This allows determining the percentage of people qualified for credit.

Data Analysis

Type : CREATION_INTERNE

The examples use data (mycaslib.creditqualify) that is assumed to exist in the CAS library 'mycaslib'. The first step creates a new table (mycaslib.qualifyapps) from it, to be used in the second step. To make the example self-contained, the 'mycaslib.creditqualify' table should be created or be a SASHELP dataset.

1 Code Block

DATA STEP Data

Explanation :
This DATA step code creates a new table named `mycaslib.qualifyapps`. It reads data from `mycaslib.creditqualify` and adds a `Count` variable. If 'Credit_Qualification' is 'N/A', `Count` is set to 0; otherwise, `Count` is set to 1.

Copied!

1	DATA mycaslib.qualifyapps;
2	SET mycaslib.creditqualify;
3	IF Credit_Qualification='N/A' THEN Count=0;
4	ELSE Count=1;
5	RUN;

2 Code Block

PROC CAS

Explanation :
This section uses the `freqTab.freqTab` action of PROC CAS to generate a frequency table. It takes the `qualifyapps` table created previously, uses the `Count` variable as a weight, and calculates frequencies for the `Credit_Qualification` variable. This allows obtaining the distribution of credit qualifications.

Copied!

1	PROC CAS;
2	ACTION freqTab.freqTab/
3	TABLE='qualifyapps',
4	weight='Count',
5	tabulate={'Credit_Qualification'};
6	QUIT;

Pro Tip

If your goal is simply to exclude rows, using the where= parameter directly inside the CAS action is often more memory-efficient than creating a new physical column like Count.

This material is provided "as is" by We Are Cas. There are no warranties, expressed or implied, as to merchantability or fitness for a particular purpose regarding the materials or code contained herein. We Are Cas is not responsible for errors in this material as it now exists or will exist, nor does We Are Cas provide technical support for it.