Introduction to SAS Software

Introduction to SAS

• SAS is a software package developed by SAS Institute.
• It's used for data analysis, statistical analysis, business intelligence, and decision-making.
• SAS is a programming language.
• SAS provides a graphical user interface (GUI) for interacting with applications.
• The GUI enables users to create reports, graphs, data summaries, and statistical analyses.
• SAS is platform-independent, so it can run on various operating systems like Linux and Windows.

Data Source

• Data is stored in electronic files and software files.
• Software data is accessible up to 52 electronic files, also referred to as database files.
• Some examples include MS Access, Excel, SQL Server, and Oracle DB Server.

Data Consolidation

• This involves collecting and integrating data from multiple sources into a single destination.
• Different data sources are combined into a single data store.

Data Warehouse

• A data warehouse (DW) is a repository of data derived from various sources.
• It's a relational database designed for data consolidation, analysis, and reporting at different aggregate levels.

Data Provisioning

• This is for accessing, analyzing, and reporting on data used in various applications.

History of SAS

• James Goodnight, a graduate student in statistics, developed the initial SAS programming tool in 1959.
• SAS was launched in New York in 1970 as statistical analysis tools.
• In 1976, Jim Goodnight continued working on SAS and founded SAS Institute.
• In 1980, SAS evolved into a software package for generating graphical reports.
• In 1990, SAS expanded functionalities to encompass data warehousing.
• Currently, SAS Institute is the world's largest privately held software company, and Goodnight serves as its CEO.

SAS Components

• SAS has over 200 components available.
• Components include (but aren't limited to) basic procedures for data management, statistical analysis, graphical presentation, operations research, econometrics, interactive matrix language, application facility, quality control, clinical trial analysis, data mining, and business intelligence applications.

SAS Window Environment

• SAS Window software aids programmers by providing tools to reduce code writing time.
• The SAS Window environment has five primary components: Project Tree, Menu, Search Box, Editor Window, and Log Window.
• The Project Tree manages projects and files.
• The Menu provides the list of commands for actions.
• The Search Box facilitates searching within programs and files.
• The Editor Window is the primary area for writing and editing SAS code.
• The Log Window records session actions, errors, and warnings.

Libraries in SAS

• Libraries in SAS are storage locations for data sets.
• Libraries can be created and saved in SAS.
• Each SAS library's name is restricted to 8 characters.
• The default libraries include SASHELP (for sample data sets and information for controlling a SAS session), SASUSER (for user-specific data, programs, and files), and WORK (for temporary storage of data sets without explicit library specification).

SAS Programs

• SAS programs are sequences of statements executed in order.
• Each statement provides instructions to the SAS system.
• SAS programs can have DATA steps and PROC steps.
• DATA steps manage tables and data.
• PROC steps handle data analysis and reporting using built-in procedures.

Writing DATA and PROC Blocks

• To specify or declare dataset on PROC and variables in a program, specific rules must be followed.
• SAS filename and variable names have length restrictions (upto 32 characters).
• Variable names start with letters or underscores and can contain letters, numbers, and underscores.
• Specific rules or syntax formats need to followed in SAS.

Raw Data Types

• Internal: Data treated directly by SAS as internal data and are accessed and processed by using CARDS or DATALINES keywords.
• External: Data in external files accessed by the INFILE keyword and paths placed in double quotes

SAS Input Statements

• INPUT statements in SAS guide how to structure raw data tables and read data.
• SAS input statements categorize data reading using six styles: LIST, COLUMN, FORMATTED, NAME, ABSOLUTE, and MIXED. These styles provide different ways to organize and extract data from input sources.

SAS Data Reading Techniques

• Column Pointer: The @n option allows you to instruct SAS to skip a specific number of characters within a line for certain variables.
• Line Pointer: Forward slash (/) refers to a line location, while #n specifies the precise line number for a pointer shift.
• Colon Modifier: Use this to read data lines exceeding certain character length.
• Ampersand Modifier: Used to manage blank or other delimiters within variable values.
• Tilde Modifier: This allows for single and multiple quotation marks, and delimiters to be kept within a character variable value.

Data Trailing

• Double Trailing: Used when multiple observations are present, this option is specified at the end of the INPUT statement enabling SAS to manage multiple data observations on a single line.
• Single Trailing: Used when you want to read some part of the raw data, this option is specified after a particular variable in the INPUT statement.

INFILE Statement

• This precedes the INPUT statement informing SAS of the location and path of the data file.
• Path names are surrounded by double quotes. This statement helps read external files (text files, for example).

INFILE Options

• Options are used for directing the data reading process like;
  • FIRSTOBS: Specifies the starting line number for data reading.
  • OBS: Specifies the ending data line number for reading.
  • MISSOVER: Handles missing data values and assigns missing values to any remaining variables.
  • DLM (Delimiter): Reads data values separated by a specified character.
  • DSD (Data Separated Delimiter) Separates data values based on columns.
  • FLOWOVER: Enables processing of data regardless of whether errors are encountered in the data stream.
  • STOPOVER: Controls how SAS handles errors or missing parts in the data stream.
  • TRUNCOVER: Adjusts the variable lengths to manage and organize effectively.

Additional SAS Methods

• TRUNCOVER: Adjusts variable lengths to fit the data.
• SCANOVER: Searches for a string to assist with character reading.
• LRCEL (Long Record Length): Adjusts the length of records.

IF-THEN Conditional Statements

• Used for conditional statements like subsetting and conditional logic.
• Used to create new variables, delete observations, or change existing variable values.

Iterative DO Loops

• For repetitive calculations or tasks in a dataset.

DO WHILE Loops

• Loops that continues when condition is true.

DO UNTIL Loops

• Loops that continues when the condition is false.

WHERE Statement (Expression)

• Used to create a dataset subset for storage or reporting.
• WHERE statement should be used in a data step when you are performing SET, MERGE, or UPDATE statement.

FORMAT Statement

• Used for transforming data for formatting or reporting purposes.
• Formats are categorized as temporary if written in PROC blocks or permanent if written in DATA blocks.

PROC SORT

• Used for sorting data in a dataset.
• Specifies ascending or descending order in BY statement

NODUP and NODUPKEY

• NODUP: Removes all but the first of duplicate observations.
• NODUPKEY: Removes duplicate observations based on the BY variable.