The SAS dataset contains values that are organized as rows and columns that can be processed (read/written) by SAS. The dataset can be a data file (table) or view. Either way, a dataset is typically rectangular in format. The dataset has a descriptor portion and data portion. While in the following table, we can only see the column/variable names, the descriptor portion holds further information such as the number of rows (more commonly referred to as observations) in the dataset, date and time of creation, and the operating environment in which it was created. This section is called the data portion, which holds all the data values:

The maximum number of observations that can be counted for a SAS dataset is determined by the long integer data type size for the operating environment. In operating environments with a 32-bit long integer, the maximum number...

We used code in the first program to create an output dataset. The dataset was created by what is known as data steps. Data steps are a collection of statements that can help create, modify, and control the output. SAS also leverages Structured Query Language (SQL). Let's review the basic syntax for data steps and SQL within SAS. We will continue to explore more advanced versions of code throughout this book.

The following code represents one of the simplest forms of a data step:

DATA WORK.Air;
SET SASHELP.Air;
RUN;

Using the SET statement in this data step, we have specified the dataset that we want to refer to. There are no conditional statements in the program...

The SAS LOG section of your coding environment is where all the actions performed that have been by the user in the current session are stored. These include instances of program submission and also messages about any programs that you might have terminated while they were executing. Apart from these, the SAS LOG also contains system-generated messages. These are of two types. The first instance is where the SAS version and a few other details about your system are written to the LOG. The second is when responses to the user code are generated and written to the LOG. The response could state that the program has run successfully, failed, or has some syntax issues that have been ignored. The responses are categorized in NOTE, INFO, WARNING, and ERROR categories. Program submission messages can be easily identified in the LOG as they have a line number associated with them...

There are many built-in SAS options that apply to the dataset. The broad purpose of these options is to help us do the following:

• Rename variables
• Select variables for subsetting
• Retain select variables
• Specify the password for a dataset, compress it, and encrypt it

Throughout this book, we will be looking at various dataset options. We will begin by exploring the compress, encrypt, and index options.

Compression reduces the number of bytes that are required to store each observation. The advantage of compression is that it requires less storage, owing to the fact that fewer bytes are required and fewer I/O operations are necessary to read or write to the data during processing.

The biggest disadvantage...

An SAS operator is a symbol that represents a comparison, arithmetic calculation, or logical operation. We will look into the various operator in the following sections.

For performing arithmetic operations, the following can be used in SAS:

Comparison operators set up a comparison, operation, or calculation with two variables, constants, or expressions. If the comparison is true, the result...

There are usually three scenarios when formats are used in SAS:

• To format the data available to make it more readable
• To assign a specific format to make the data more meaningful
• To alter the data type to make calculations, derivations, and so on

SAS stores all dates as single unique numbers in a numeric format. All dates are stored as the number of dates from January 1, 1960. If you think of a number line, all dates prior to January 1, 1960 are negative and all dates after that are positive. It’s cumbersome to decode the numeric dates. An easier alternative is to get SAS to display a date using a date format. Let's look at our cost of living dataset with the addition...

In many instances, we will need to select only a proportion of the overall dataset. This selection or subsetting of the dataset can be done in three ways:

• Use of the WHERE and IF statements
• Using the SAS dataset OPTIONS
• Using the DROP and KEEP statements

Let's only select the data where the records have been updated in 2019:

DATA Updated_2019;
SET Cost_Living;
WHERE Year(Updated) = 2019;
RUN;

The following output will be generated:

In this instance, we could have also used the IF statement to generate similar output. We can also use WHERE and IF statements in the same data step:

DATA Updated_2019;
SET Cost_Living;
WHERE Year(Updated) = 2019;
IF Index >= 80; 
RUN;

This further...

Some SAS software packages require minimal coding experience to get your work done. However, once you want to do anything advanced, it gets restrictive to use just the predefined options available in the package. At that moment, most users tend to write code. Chances are that if you want to go beyond what SAS already offers as predefined procedures and functionalities, you want to write advanced programs/macros. While we will delve into macros later on in this book, there is an aspect of SAS that is being leveraged less by some users. These are the dictionaries. But what are they and how can they be useful?

Some reasons for using dictionary tables are as follows:

• They hold all the information about SAS libraries, datasets, macros, and external files that are being used in the SAS session.
• It is a read-only view, so there are no chances of it being compromised...

In this chapter, we covered the basics of SAS dataset fundamentals and how to create a table, compile it, and then execute it. We also looked into the basic syntax of the SAS programming language. Then, we learned how to compress, encrypt, and index a dataset. Finally, we learned the various operators in SAS, and how to format and subset the dataset.

In the next chapter, we will learn how to manipulate, sanitize, and transform data.