User manual SAS SAS.9.1.3 LANGUAGE REFERENCE DICTIONARY

SAS Publishing ® SAS 9.1.3 Language Reference: Dictionary ® Fifth Edition The correct bibliographic citation for this manual is as follows: SAS Institute Inc. 2006. SAS ® 9.1.3 Language Reference: Dictionary, Fifth Edition, Volumes 1-4. Cary, NC: SAS Institute Inc. SAS® 9.1.3 Language Reference: Dictionary, Fifth Edition, Volumes 1-4 Copyright © 2002-2006, SAS Institute Inc., Cary, NC, USA ISBN-13: 978-1-59994-098-4 ISBN-10: 1-59994-098-1 All rights reserved. Produced in the United States of America. For a hard-copy book: No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, or otherwise, without the prior written permission of the publisher, SAS Institute Inc. For a Web download or e-book: Your use of this publication shall be governed by the terms established by the vendor at the time you acquire this publication. U.S. Government Restricted Rights Notice. Use, duplication, or disclosure of this software and related documentation by the U.S. government is subject to the Agreement with SAS Institute and the restrictions set forth in FAR 52.227-19 Commercial Computer Software-Restricted Rights (June 1987). SAS Institute Inc., SAS Campus Drive, Cary, North Carolina 27513. 1st printing, July 2006 2nd printing, August 2006 SAS® Publishing provides a complete selection of books and electronic products to help customers use SAS software to its fullest potential. For more information about our e-books, e-learning products, CDs, and hard-copy books, visit the SAS Publishing Web site at support.sas.com/pubs or call 1-800-727-3228. SAS® and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of SAS Institute Inc. in the USA and other countries. ® indicates USA registration. Other brand and product names are registered trademarks or trademarks of their respective companies. Contents What's New vii Overview vii SAS System Features viii SAS Language Elements x PART 1 Dictionary of Language Elements Chapter 1 The SAS Language Reference: Dictionary 4 Introduction to the SAS 9.1 Language Reference: Dictionary 3 1 3 Chapter 2 Definition of Data Set Options Syntax 6 Using Data Set Options 6 Data Set Options by Category Dictionary 9 4 SAS Data Set Options 6 7 5 Chapter 3 Definition of Formats 73 Syntax 74 Using Formats 74 Byte Ordering for Integer Binary Data on Big Endian and Little Endian Platforms Data Conversions and Encodings 79 Working with Packed Decimal and Zoned Decimal Data 80 Formats by Category 84 Dictionary 95 4 Formats 69 77 Chapter 4 Definitions of Functions and CALL Routines 268 Syntax 269 Using Functions 271 Using Random-Number Functions and CALL Routines 273 Pattern Matching Using SAS Regular Expressions (RX) and Perl Regular Expressions (PRX) 276 Base SAS Functions for Web Applications 286 Functions and CALL Routines by Category 286 Dictionary 310 References 1005 4 Functions and CALL Routines 259 Chapter 5 Definition of Informats 1010 Syntax 1010 Using Informats 1011 4 Informats 1007 iv Byte Ordering for Integer Binary Data on Big Endian and Little Endian 1013 Platforms Working with Packed Decimal and Zoned Decimal Data Informats by Category 1019 Dictionary 1026 1015 Chapter 6 Definition of ARM Macros 1137 Using ARM Macros 1138 Defining User Metrics in ARM Macros 1145 Defining Correlators in ARM Macros 1146 Enabling ARM Macro Execution Setting the Macro Environment 1147 1149 4 SAS ARM Macros 1137 Using ARM Post-Processing Macros 1150 Troubleshooting Error Messages 1151 ARM Macros by Category 1152 Dictionary 1153 Chapter 7 Definition of Statements 1174 DATA Step Statements 1174 Global Statements Dictionary 1184 1179 4 Statements 1171 Chapter 8 Definition of System Options 1553 Syntax 1553 Using SAS System Options 1553 Comparisons 1558 SAS System Options by Category Dictionary 1568 1559 4 SAS System Options 1549 PART 2 Appendixes Appendix 1 The DATA Step Component Object Interface 4 DATA Step Object Attributes and Methods 1765 1766 1763 1765 Dot Notation and DATA Step Component Objects Dictionary 1767 Appendix 2 Introduction 4 DATA Step Debugger 1794 1793 Basic Usage 1795 Advanced Usage: Using the Macro Facility with the Debugger Examples 1797 Commands Dictionary 1809 1810 1796 v Appendix 3 Recommended Reading 4 SAS Utility Macro 1827 Appendix 4 4 Recommended Reading 1831 1831 Index 1833 vi vii What's New Overview New and enhanced features in Base SAS save you time, effort, and system resources by providing faster processing and easier data access and management, more robust analysis, and improved data presentation. 3 By using new SAS system options that enable threading and the use of multiple CPUs, the following SAS procedures take advantage of multi-processing I/O: SORT, SQL, MEANS, TABULATE, and REPORT. 3 The LIBNAME statement now supports secure access to SAS libraries on a WebDAV server. 3 You can now use longer, easier-to-read names for user-created formats and informats. See "Rules for Words and Names in the SAS Language" in SAS Language Reference: Concepts. 3 Two pre-defined component objects for the DATA step enable you to quickly store, search, and retrieve data based on lookup keys. 3 The FILENAME statement now supports directory services, multiple FTP service commands, and Secure Sockets Layering (SSL). 3 The Application Response Measurement (ARM) system enables you to monitor the availability and performance of transactions within and across diverse applications. 3 The Perl regular expression (PRX) functions and CALL routines use a modified version of Perl as a pattern-matching language to enhance search-and-replace operations on text. 3 New character functions search and compare character strings in addition to concatenating character strings. 3 There are several new descriptive statistic functions and mathematical functions. 3 New formats, informats, and functions support international and local values for money, datetime, and Unicode values. All data set options, formats, informats, functions, and system options that relate to national language support are documented in the new SAS National Language Support (NLS): User's Guide. 3 A new ODS statement enables you to render multiple ODS output formats without re-running a PROC or a DATA step. See the SAS Output Delivery System: User's Guide. viii What's New Note: 3 This section describes the features of Base SAS that are new or enhanced since SAS 8.2. 3 z/OS is the successor to the OS/390 operating system. SAS 9.1 (and later) is supported on both OS/390 and z/OS operating systems and, throughout this document, any reference to z/OS also applies to OS/390, unless otherwise stated. 4 SAS System Features Application Response Measurement (ARM) Application Response Measurement (ARM) enables you to monitor the availability and performance of transactions within and across diverse applications. The SAS ARM interface consists of the implementation of the ARM API as ARM macros and an ARM agent. An ARM agent generates calls to the ARM macros. New ARM system options enable you to manage the ARM environment and to log internal SAS processing transactions. See "Monitoring Performance Using Application Response Measurement (ARM)" in SAS Language Reference: Concepts, "ARM Macros" on page xx, and "System Options" on page xxi. Cross-Environment Data Access (CEDA) CEDA processes SAS files that were created on a different host. This is especially useful if you have upgraded from a 32­bit platform to a 64­bit platform. Messages in the SAS log notify you when CEDA is being used to process a SAS file. See "Processing Data Using Cross-Environment Data Access (CEDA)" in SAS Language Reference: Concepts. DATA Step Object Attributes and Methods SAS now provides two pre-defined component objects for use in a DATA step: the hash object and the hash iterator object. These objects enable you to quickly and efficiently store, search, and retrieve data based on lookup keys. The DATA step component object interface enables you to create and manipulate these component objects by using statements, attributes, and methods. You use the DATA step object dot notation to access the component object's attributes and methods. The hash and hash iterator objects have one attribute, fourteen methods, and two statements associated with them. See Appendix 1, "DATA Step Object Attributes and Methods," on page 1765. Engines 3 The default BASE engine in SAS supports longer format and informat names, thread-enabled procedures such as the SORT and SUMMARY procedures, and more than 32,767 variables in a SAS data set. 3 The metadata LIBNAME engine enables you to use metadata in order to access and augment data that is identified by the metadata. The metadata engine What's New ix 3 3 3 3 retrieves information about the target SAS data library from metadata objects in a specified SAS Metadata Repository on the SAS Metadata Server. The metadata engine provides a consistent method for accessing many data sources. That is, SAS provides different engines that have different options, behavior, and tuning requirements. By taking advantage of metadata, the necessary information that is required to access data can be created in one central location so that applications can use the metadata engine to access different sources of data, without having to understand the differences and details of each SAS engine. See the SAS Metadata LIBNAME Engine: User's Guide. The XML LIBNAME engine imports and exports a broader variety of XML documents. The XMLMAP= option specifies a separate XML document that contains specific XMLMap syntax. The XMLMap syntax, Version 1.2, tells the XML engine how to interpret the XML markup in order to successfully import an XML document. See the SAS Metadata LIBNAME Engine: User's Guide. The new SASEDOC LIBNAME engine enables you to bind output objects that persist in an ODS document. See the SAS Output Delivery System: User's Guide. The new SAS Information Maps LIBNAME Engine provides a read-only way to access data that is generated from a SAS Information Map and to bring it into a SAS session. After you retrieve the data, you can run almost any SAS procedure against it. See the Base SAS Guide to Information Maps. The new character variable padding (CVP) engine expands character variable lengths, using a specified expansion amount, so that character data truncation does not occur when a file requires transcoding. Character data truncation can occur when the number of bytes for a character in one encoding is different from the number of bytes for the same character in another encoding, such as when a single-byte character set (SBCS) is transcoded to a double-byte character set (DBCS). See the SAS National Language Support (NLS): User's Guide. Indexing When creating an index that requires sorting, SAS tries to sort the data by using the thread-enabled sort. By dividing the sorting task into separately executable processes, the time that is required to sort the data can be reduced. See the topic "Creating an Index" in Understanding SAS Indexes in SAS Language Reference: Concepts. Integrity Constraints Variables in a SAS data file can now be part of both a primary key (general integrity constraint) and a foreign key (referential integrity constraint). However, there are restrictions when defining a primary key constraint and a foreign key constraint that use the same variables. See the topic "Overlapping Primary Key and Foreign Key Constraints" in Understanding Integrity Constraints in SAS Language Reference: Concepts. Restricted System Options System administrators can restrict system options from being modified by a user. You can use the RESTRICT option in the OPTIONS procedure to list the restricted options. The implementation of restricted options is specific to the operating environment. For details about how to restrict options, see the configuration guide for your operating environment. For information about listing restricted options, see the OPTIONS procedure in the Base SAS Procedures Guide. x What's New SAS Utility Macro The SAS utility macro, %DS2CSV, is available now in Base SAS. This macro converts SAS data sets to comma-separated values (CSV) files. Prior to SAS 9.1, this macro was available only for SAS/IntrNet users. Universal Unique Identifiers A Universal Unique Identifier (UUID) is a 128-bit identifier that consists of date and time information, and the IEEE node address of a host. UUIDs are useful when objects such as rows or other components of a SAS application must be uniquely identified. For more information, see "Universal Unique Identifiers" in SAS Language Reference: Concepts. SAS Language Elements Descriptions of the new and enhanced language elements for national language support can be found in "What's New for SAS 9.0 and 9.1 National Language Support" in the SAS National Language Support (NLS): User's Guide. Data Set Options 3 The following data set options are new: OBSBUF= determines the size of the view buffer for processing a DATA step view. SPILL= specifies whether to create a spill file for non-sequential processing of a DATA step view. 3 The following data set options are enhanced: BUFNO= supports the same syntax as the BUFNO= system option in order to specify the number of buffers to be allocated for processing a SAS data set. BUFSIZE= supports the same syntax as the BUFSIZE= system option in order to specify the permanent buffer page size for an output SAS data set. FIRSTOBS= supports the same syntax as the FIRSTOBS= system option in order to specify which observation SAS processes first. OBS= supports the same syntax as the OBS= system option in order to specify when to stop processing observations. Formats 3 The maximum length for character format names is increased to 31. The maximum length for numeric format names is increased to 32. What's New xi 3 Several formats have been enhanced with default and range values. 3 The following formats are new: MMYY writes date values in the form mmMyy, where M is the separator and the year is written in either 2 or 4 digits. PERCENTN produces percentages, using a minus sign for negative values. YYMM writes date values in the form yyMmm, where the year is written in either 2 or 4 digits and M is the separator. YYQ writes date values in the form yyQq, where the year is written in either 2 or 4 digits, Q is the separator, and q is the quarter of the year. YYQR writes date values in the form yyQqr, where the year is written in either 2 or 4 digits, Q is the separator, and qr is the quarter of the year expressed in Roman numerals. 3 The PVALUE format now returns missing values that are specified by the MISSING= system option. Functions and CALL Routines New functions and CALL routines include character, mathematical, descriptive statistical, and special functions, and character-string matching functions that can use PERL expressions. 3 The following character functions are new: ANYALNUM searches a character string for an alphanumeric character and returns the first position at which it is found. ANYALPHA searches a character string for an alphabetic character and returns the first position at which it is found. ANYCNTRL searches a character string for a control character and returns the first position at which it is found. ANYDIGIT searches a character string for a digit and returns the first position at which it is found. ANYFIRST searches a character string for a character that is valid as the first character in a SAS variable name under VALIDVARNAME=V7, and returns the first position at which that character is found. ANYGRAPH searches a character string for a graphical character and returns the first position at which it is found. ANYLOWER searches a character string for a lowercase letter and returns the first position at which it is found. xii What's New ANYNAME searches a character string for a character that is valid in a SAS variable name under VALIDVARNAME=V7, and returns the first position at which that character is found. ANYPRINT searches a character string for a printable character and returns the first position at which it is found. ANYPUNCT searches a character string for a punctuation character and returns the first position at which it is found. ANYSPACE searches a character string for a white-space character (blank, horizontal tab, vertical tab, carriage return, line feed, or form feed), and returns the first position at which it is found. ANYUPPER searches a character string for an uppercase letter and returns the first position at which it is found. ANYXDIGIT searches a character string for a hexadecimal character that represents a digit and returns the first position at which that character is found. CAT concatenates character strings without removing leading or trailing blanks. CATS concatenates character strings and removes leading and trailing blanks. CATT concatenates character strings and removes trailing blanks only. CATX concatenates character strings, removes leading and trailing blanks, and inserts separators. CHOOSEC returns a character value that represents the results of choosing from a list of arguments. CHOOSEN returns a numeric value that represents the results of choosing from a list of arguments. COMPARE returns the position of the left-most character by which two strings differ, or returns 0 if there is no difference. COMPGED compares two strings by computing the generalized edit distance. COMPLEV compares two strings by computing the Levenshtein edit distance. COUNT counts the number of times that a specific substring of characters appears within a character string that you specify. What's New xiii COUNTC counts the number of specific characters that either appear or do not appear within a character string that you specify. FIND searches for a specific substring of characters within a character string that you specify. FINDC searches for specific characters that either appear or do not appear within a character string that you specify. IFC returns a character value that matches an expression. IFN returns a numeric value that matches an expression. LENGTHC returns the length of a character string, including trailing blanks. LENGTHM returns the amount of memory (in bytes) that is allocated for a character string. LENGTHN returns the length of a non-blank character string, excluding trailing blanks, and returns 0 for a blank character string. NLITERAL converts a character string that you specify to a SAS name literal (n-literal). NOTALNUM searches a character string for a non-alphanumeric character and returns the first position at which it is found. NOTALPHA searches a character string for a non-alphabetic character and returns the first position at which it is found. NOTCNTRL searches a character string for a character that is not a control character and returns the first position at which it is found. NOTDIGIT searches a character string for any character that is not a digit and returns the first position at which that character is found. NOTFIRST searches a character string for an invalid first character in a SAS variable name under VALIDVARNAME=V7, and returns the first position at which that character is found. NOTGRAPH searches a character string for a non-graphical character and returns the first position at which it is found. NOTLOWER searches a character string for a character that is not a lowercase letter and returns the first position at which that character is found. xiv What's New NOTNAME searches a character string for an invalid character in a SAS variable name under VALIDVARNAME=V7, and returns the first position at which that character is found. NOTPRINT searches a character string for a non-printable character and returns the first position at which it is found. NOTPUNCT searches a character string for a character that is not a punctuation character and returns the first position at which it is found. NOTSPACE searches a character string for a character that is not a white-space character (blank, horizontal tab, vertical tab, carriage return, line feed, or form feed), and returns the first position at which it is found. NOTUPPER searches a character string for a character that is not an uppercase letter and returns the first position at which that character is found. NOTXDIGIT searches a character string for a character that is not a hexadecimal digit and returns the first position at which that character is found. NVALID checks a character string for validity for use as a SAS variable name in a SAS statement. PROPCASE converts all words in an argument to proper case. PRXCHANGE performs a pattern-matching replacement. PRXPOSN returns the value for a capture buffer. SCANQ th returns the n word from a character expression and ignores delimiters that are enclosed in quotation marks. STRIP returns a character string with all leading and trailing blanks removed. SUBPAD returns a substring that has a length you specify, using blank padding if necessary. SUBSTRN returns a substring that allows a result with a length of 0. What's New xv 3 The following descriptive statistics functions are new: GEOMEAN returns the geometric mean. GEOMEANZ returns the geometric mean without fuzzing the values of the arguments that are approximately 0. HARMEAN returns the harmonic mean. HARMEANZ returns the harmonic mean without fuzzing the values of the arguments that are approximately 0. IQR returns the interquartile range. LARGEST th returns the k largest non-missing value. MAD returns the median absolute deviation from the median. MEDIAN computes median values. MODZ returns the remainder from the division of the first argument by the second argument; uses 0 fuzzing. PCTL computes percentiles. RMS returns the root mean square. SMALLEST th returns the k smallest non-missing value. 3 The following External Files function is new: DCREATE creates an external directory. 3 The following macro functions are new: SYMEXIST indicates the existence of a macro variable. SYMGLOBL indicates whether a macro variable has global scope in the DATA step during DATA step execution. SYMLOCAL indicates whether a macro variable has local scope in the DATA step during DATA step execution. 3 The following mathematical functions are new: BETA returns the value of the beta function. COALESCE returns the first non-missing value from a list of numeric arguments. xvi What's New COALESCEC returns the first non-missing value from a list of character arguments. LOGBETA returns the logarithm of the beta function. 3 The following probability function is new: LOGCDF returns the logarithm of a left cumulative distribution function. 3 The following quantile function is new: QUANTILE returns the quantile from the specified distribution. 3 The following special function is new: UUIDGEN returns the short or the binary form of a Universal Unique Identifier (UUID). 3 The following state and ZIP code function is new: ZIPCITY returns a city name and the two-character postal code that corresponds to a ZIP code. 3 The following trigonometric function is new: ATAN2 returns the arc tangent of two numeric variables. 3 The following truncation functions are new: CEILZ returns the smallest integer that is greater than or equal to the argument; uses 0 fuzzing. FLOORZ returns the largest integer that is less than or equal to the argument; uses 0 fuzzing. INTZ returns the integer portion of the argument; uses 0 fuzzing. ROUND rounds the first argument to the nearest multiple of the second argument, or to the nearest integer when the second argument is omitted. ROUNDE rounds the first argument to the nearest multiple of the second argument, and returns an even multiple when the first argument is halfway between the two nearest multiples. ROUNDZ rounds the first argument to the nearest multiple of the second argument; uses 0 fuzzing. What's New xvii 3 The following variable information functions are new: VVALUE returns the formatted value that is associated with the variable that you specify. VVALUEX returns the formatted value that is associated with the argument that you specify. 3 Using Perl regular expression (PRX) functions and CALL routines is new. The following PRX functions are new. For more information, see "Pattern Matching Using SAS Regular Expressions (RX) and Perl Regular Expressions (PRX)" on page 276. PRXMATCH searches for a pattern match and returns the position at which the pattern is found. PRXPAREN returns the last bracket match for which there is a match in a pattern. PRXPARSE compiles a Perl regular expression that can be used for pattern-matching a character value. CALL PRXCHANGE performs a pattern-matching substitution. CALL PRXDEBUG enables Perl regular expressions in a DATA step to send debug output to the SAS log. CALL PRXFREE frees unneeded memory that was allocated for a Perl regular expression. CALL PRXNEXT returns the position and length of a substring that matches a pattern and iterates over multiple matches within one string. CALL PRXPOSN returns the start position and length for a capture buffer. CALL PRXSUBSTR returns the position and length of a substring that matches a pattern. 3 The following CALL routines are new: CALL ALLPERM generates all permutations of the values of several variables. CALL CATS concatenates character strings and removes leading and trailing blanks. CALL CATT concatenates character strings and removes trailing blanks only. CALL CATX concatenates character strings, removes leading and trailing blanks, and inserts separators. CALL COMPCOST sets the costs of operations for later use by the COMPGED function. xviii What's New CALL LOGISTIC returns the logistic value of each argument. CALL MISSING assigns a missing value to specified character or numeric variables. CALL RANPERK randomly permutes the values of the arguments and returns a permutation of k out of n values. CALL RANPERM randomly permutes the values of the arguments. CALL SCAN returns the position and length of a given word in a character expression. CALL SCANQ returns the position and length of a given word in a character expression, and ignores delimiters that are enclosed in quotation marks. CALL SOFTMAX returns the softmax value for each argument. CALL STDIZE standardizes the values of one or more variables. CALL STREAMINIT specifies a seed value to use for subsequent random number generation by the RAND function. CALL SYMPUTX assigns a value to a macro variable and removes both leading and trailing blanks. CALL TANH returns the hyperbolic tangent of each argument. CALL VNEXT returns the name, type, and length of a variable that is used in a DATA step. What's New xix 3 The following functions are enhanced: COMPRESS accepts a third optional argument that can modify the characters in the second argument. EXIST accepts all SAS data library type members. A third optional argument enables you to specify a generation data set number. INDEXW accepts a third optional argument that enables you to use delimiters for inter-word boundaries. accepts an argument that enables you to use a Perl regular expression. SUBSTR (left of=) assigns a length of 8 to an undeclared variable when the function is compiled. Informats 3 The maximum length for character informat names is increased to 30. The maximum length for numeric informat names is increased to 31. 3 The following informats are new: ANYDTDTE extracts date values from DATE, DATETIME, DDMMYY, JULIAN, MMDDYY, MONYY, TIME, or YYQ informat values. ANYDTDTM extracts datetime values from DATE, DATETIME, DDMMYY, JULIAN, MMDDYY, MONYY, TIME, or YYQ informat values. ANYDTTME extracts time values from DATE, DATETIME, DDMMYY, JULIAN, MMDDYY, MONYY, TIME, or YYQ informat values. STIMERw. reads time values and determines whether the values are hours, minutes, or seconds; reads the output of the STIMER system option.