SAS Base Programming flashcards

32 cards. Read the term and the definition here. The deck at the top of this page is optional.

LIBNAME statement

Assigns a libref to a directory or SAS library so data sets can be referenced as libref.member. Syntax: LIBNAME libref 'path'; librefs are 1-8 characters.

Program Data Vector

The DATA step's memory area holding one observation's variables during execution. It is created at compile time and reset each iteration except for retained variables.

INFILE DSD option

Reads delimited files, treating consecutive delimiters as missing values and stripping quotation marks from character values. Used with DLM= to set the delimiter.

MISSOVER option

INFILE option that assigns missing values to any variables not filled by a short input record instead of holding the record for more data.

PROC IMPORT DBMS=

Reads external files such as XLSX, CSV, or delimited text into a SAS data set. DBMS= specifies the file type and OUT= names the output data set.

ARRAY statement

Defines a group of variables referenced by a common name and index, for example ARRAY a[3] x1-x3; DIM(a) returns the number of elements.

Automatic variable _N_

Counts DATA step iterations. It equals 1 during the first iteration and increases by one each subsequent iteration; useful for selecting rows.

MERGE versus SET

SET stacks observations from multiple data sets vertically. MERGE combines them horizontally by matching BY values and requires sorted input for one-to-one matching.

PROC SORT NODUPKEY

Sorts a data set and removes observations with duplicate values of the BY variables, keeping the first occurrence of each key.

RETAIN statement

Prevents a variable from being reset to missing at the start of each DATA step iteration and can assign an initial value.

WHERE versus IF

WHERE selects observations before they enter the PDV and cannot use newly computed variables. IF executes after the PDV is populated and can use computed values.

LAG function

Returns the value of its argument from a previous DATA step iteration; LAG(n) looks back n iterations using an internal queue.

FIRST. and LAST. flags

Automatic variables that are 1 for the first or last observation in a BY group. They require data sorted by the BY variables.

PROC TRANSPOSE

Converts observations to variables or variables to observations. VAR selects columns, ID names new columns, and BY preserves group structure.

SUM statement

Syntax var+expression; retains var across iterations, ignores missing values in the expression, and treats them as zero.

INPUT and PUT functions

INPUT converts a character value to numeric using an informat. PUT converts a numeric value to character using a format.

_ERROR_ variable

Automatic DATA step variable set to 1 when a data error occurs during execution and reset to 0 at the start of each iteration.

ERRORABEND option

System option that causes SAS to terminate immediately when an error occurs. The default NOERRORABEND allows processing to continue.

OPTIONS ERRORS=

Limits how many observations with data errors print error messages in the log. Default is 20; setting it to 0 suppresses messages.

PUT _ALL_ statement

Writes every variable currently in the PDV, with names and values, to the SAS log. Used to debug DATA step logic.

ABORT statement

Stops DATA step execution. ABORT CANCEL cancels submitted statements, while ABORT RETURN returns control to the calling environment.

SYSERR macro variable

Contains the return code from the most recent SAS procedure or DATA step; a nonzero value indicates an error condition.

Compile versus execution errors

Compile errors are syntax problems detected before the DATA step runs. Execution errors occur while processing data, such as invalid numeric conversion.

Missing values note

Log message 'Missing values were generated' appears when numeric operations create missing values, often from character-to-numeric conversion.

PROC PRINT NOOBS

Prints a data set without the observation number column. Other options include LABEL, BY, ID, VAR, and SUM.

PROC FREQ MISSING

Includes missing values as a category in frequency tables. Without MISSING, observations with missing values are excluded from the table.

PROC MEANS CLASS

Produces summary statistics for each level of the CLASS variables. NWAY outputs only the highest-level class combination.

PROC REPORT DEFINE

Specifies a column's usage, such as DISPLAY, GROUP, ORDER, or ANALYSIS, and its label, width, and format in a report.

ODS HTML statement

Routes procedure output to an HTML file. Open with ODS HTML FILE='name.html'; and close with ODS HTML CLOSE.

TITLE statement

Adds a title line above report output. TITLE; clears all titles; TITLE2 adds a second title line.

PROC TABULATE

Creates dimensional summary tables using TABLE statements with row and column expressions; CLASS variables define categories and VAR variables supply analysis values.

DATA _NULL_ PUT

A DATA step with no output data set, using PUT to write custom report lines to the log or to a file named in a FILE statement.