Introduction to SAS Statistical Package

Transcription

1 Instructor: Introduction to SAS Statistical Package Biostatistics Lecture 1 Lucy Meoni lmeoni@jhmi.edu Teaching Assistant : Sorina Eftim seftim@jhsph.edu Lecture/Lab: Room 3017 WEB site: sas@jhsph.edu to submit exercises 2 Using the PC labs SAS Version 9.0 requires basic Windows skills Set up class folder on your media (thumb drive, diskette) download files from the website for lab into class folder bring thumb drive or diskette to class Text : The Little SAS Book 3rd edition Other References: SAS online documentation Online tutor SAS system help Many, many SAS manuals SAS website WHAT IS SAS? Integrated system of software products began as software package for statistical analysis data management reporting and graphics analytic COURSE OBJECTIVES to introduce and develop skills in SAS; a statistical package used in research data analysis develop the skills necessary to create and modify a SAS data set and perform statistical analyses etc

2 COURSE TOPIC SEQUENCE introduction to SAS data definition and restructuring dates and functions file combination arrays and loops statistical procedures REFERENCE The Little SAS Book Chapter 1 and 2 WEB site: TOPICS: Introduction to SAS SAS programs SAS windowing environment SAS tables (data sets) SAS libraries Creating SAS tables IMPORT wizard StatTransfer Temporary vs permanent files 9 SAS use statements to write a series of instructions called a SAS program not command line driven (STATA) statements are written using the SAS language (a programming language that you use to manage your data). sequence of statements executed in order SAS procedures are software tools for data analysis and reporting. 10 The SAS Programming Process Debug or Modify Define the Need Create a SAS Program Enter the SAS Program Code Process the SAS Program Code Introduction to SAS Programs Review the Results 11 2

3 A SAS program is a sequence of steps that the user submits for execution. Raw Data SAS Data Set SAS Programs DATA steps are typically used to create SAS data sets. DATA Step SAS Data Set PROC Step Report PROC steps are typically used to process SAS data sets (that is, generate reports and graphs, edit data, and sort data). SAS Program LIBNAME mylib 'd:\temp\sasclass'; DATA newclass; SET mylib.class; PROC PRINT DATA=newclass; PROC MEANS DATA=newclass; CLASS gender; VAR bmi; RUN; DATA Step PROC Steps SAS steps begin with a DATA statement Step Boundaries PROC statement. SAS detects the end of a step when it encounters a RUN statement (for most steps) a QUIT statement (for some procedures) the beginning of another step (DATA statement or PROC statement). Step Boundaries : Lecture1.sas DATA newclass; SET mylib.class; PROC PRINT DATA=newclass; PROC MEANS DATA=newclass; CLASS gender; VAR bmi; RUN; Objectives Running SAS Programs Invoke the SAS System and include a SAS program into your session. Submit a program and browse the results. Navigate the SAS windowing environment. 18 3

4 SAS Windowing Environment Interactive windows enable you to interface with SAS. Navigating the SAS Windowing Environment 19 The SAS Programming Process Define the Need Create a SAS Program The SAS Programming Process Define the Need Create a SAS Program Debug or Modify Enter the SAS Program Code Process the SAS Program Code Debug or Modify Enter the SAS Program Code Process the SAS Program Code Review the Results Review the Results Objectives Entering and Executing SAS Code Enter SAS program code in the SAS windowing environment and execute the program. 24 4

5 Enter the SAS Program Code Once the planning and coding effort for a SAS program is complete, the SAS program code must be entered into the computer to process and to test the program. 25 Windowing Mode Windowing mode is a facility that enables you to enter and execute SAS programs and view the results in an interactive environment. An interactive environment permits the program to be processed immediately when submitted for execution. 26 The SAS windowing environment is made up of a collection of windows. There are three primary windows in the windowing environment The Enhanced Program Editor window enables SAS program code to be entered from the keyboard submitted for execution. SAS program elements are color-coded, including procedures, keywords, numeric and string constants, undefined keywords The Log window displays the SAS program code submitted for execution messages from SAS indicating the status of the program execution. The Output window displays reports generated by the SAS program

6 Selecting from pull-down menu Commands are used to navigate among the various windows of the SAS windowing environment and are used to execute a program. Typing the command Clicking on a tool button Depending upon the operating environment, commands can be issued by... Using function keys (F1 - F12) Entering and Executing SAS Code This demonstration illustrates entering SAS program code into the SAS Windowing Environment and executing the program. SAS Windowing Environment windowing system for editing and executing SAS programs interactive full screen collection of windows for editing programs, executing programs and displaying results five basic SAS windows Starting SAS Start SAS from the START button, PROGRAMS, The SAS System for Windows V9 system interactive full screen Enhanced Editor window, LOG window, and Explorer windows appear activate window by clicking within the window

7 ENHANCED PROGRAM EDITOR ENHANCED PROGRAM EDITOR text editor write and edit programs submit programs (use SUBMIT icon) save program statements to file with extension.sas asterisk (*) appears in title bar to indicate file has not been saved multiple windows possible 37 an ASCII editor that uses visual aides to help you write and debug your SAS programs. SAS program elements are color- coded, including procedures, keywords, informats and formats, dates, numeric and string constants, macro keywords, undefined keywords, and more. File, Open on menu to read in an existing SAS program file 38 LOG Window contains the compilation and execution results of DATA contains submitted program statements messages from SAS about compilation and execution -notes, warnings, errors save contents of this window to a file with extension.log (File, Save menu) clear window by clicking NEW icon on Toolbar LOG WINDOW CONTENTS DATA Step name of the SAS data set read and the number of observations and variables in the data set name of the SAS data set created and the number of observations and variables in the new SAS data set PROC Steps LOG WINDOW index with page numbers for successful procedures

8 OUTPUT WINDOW printable results from procedures save contents of this window to a file with extension.lst empty if program did not run - CHECK LOG window for errors indexed in the RESULTS window clear window and by clicking on NEW icon on Toolbar RESULTS WINDOW table of contents for OUTPUT window lists each part of your results in an outline form possible to save and/or print sections of results by right-clicking on section

9 EXPLORER WINDOW provides access to SAS files and libraries similar to Windows Explorer SAS FILE ORGANIZATION File Type Extension program file (EDITOR window).sas SAS table (data set).sas7bdat log file (LOG window).log Listing file (OUTPUT).lst Submitting a SAS Program When you execute a SAS program, the output generated by SAS is divided into two major parts: SAS log contains information about the processing of the SAS program, including any warning and error messages. SAS output contains reports generated by SAS procedures and DATA steps Running a SAS Program lecture1.sas This demonstration illustrates how to start a SAS session, include and submit a SAS program, and browse the results. 53 SAS Log 1 LIBNAME mylib 'd:\temp\sasclass'; NOTE: Libref MYLIB was successfully assigned as follows: Engine: V9 Physical Name: d:\temp\sasclass 2 DATA newclass; SET mylib.class; 3 4 NOTE: There were 5 observations read from the data set MYLIB.CLASS. NOTE: The data set WORK.NEWCLASS has 5 observations and 7 variables. 5 PROC PRINT DATA=newclass; 6 NOTE: There were 5 observations read from the data set WORK.NEWCLASS. 7 proc means DATA=newclass; 8 var bmi ; 9 class gender; 10 NOTE: There were 5 observations read from the data set WORK.NEWCLASS RUN; 54 9

10 PROC PRINT Output lname baseage gender race weight height BMI 1 Richardson Lowrey Tierney Sommers Kegan PROC MEANS Output The SAS System The MEANS Procedure Analysis Variable : BMI N gender Obs N Mean Std Dev ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ Minimum Maximum ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ Exercise I ONLY Mastering Fundamental Concepts 57 Objectives Define the components of a SAS data set. Define a SAS variable. Identify a missing value and a SAS date value. State the naming conventions for SAS data sets and variables. Explain SAS syntax rules. Investigate a SAS data set using the CONTENTS and PRINT procedures. 59 Data Entry Descriptor Portion SAS Data Sets External File Conversion Process SAS Data Set Other Software Files Data Portion 60 10

11 SAS data sets have a descriptor portion and a data portion. Descriptor Portion Data Portion SAS Data Sets General data set information * data set name * data set label * date/time created * storage information * number of observations Information for each variable * Name * Type * Length * Position * Format * Informat * Label Browsing the Descriptor Portion descriptor portion of a SAS data set contains general information about the SAS data set (such as data set name and number of observations) variable attributes (name, type, length, position, informat, format, label). CONTENTS procedure displays the descriptor portion of a SAS data set Browsing the Descriptor Portion General form of the CONTENTS procedure: PROC CONTENTS DATA=SAS-data-set; RUN; Partial PROC CONTENTS OUTPUT The CONTENTS Procedure Data Set Name WORK.NEWCLASS Observations 5 Member Type DATA Variables 7 Engine V9 Indexes 0 Created Friday, March 18, :02:40 PM Observation Length 64 Last Modified Friday, March 18, :02:40 PM Deleted Observations 0 Protection Compressed NO Data Set Type Sorted NO Label Alphabetic List of Variables and Attributes Example: proc contents data=work.newclass; 63 # Variable Type Len Format Informat 7 BMI Num 8 2 baseage Num 8 BEST12. F12. 3 gender Num 8 BEST12. F12. 6 height Num 8 BEST12. F12. 1 lname Char 15 $F15. $F15. 4 race Num 8 BEST12. F12. 5 weight Num 8 BEST12. F SAS Data Sets: Data Portion The data portion of a SAS data set is a rectangular table of character and/or numeric data values. lname gender BMI Richardson Lowrey Tierney Sommers Kegan Character values Numeric values Variable names Variable values 65 SAS Variable Values There are two types of variables: Character contain any value: letters, numbers, special characters, and blanks. Character values are stored with a length of 1 to 32,767 bytes. One byte equals one character. Numeric stored as floating point numbers in 8 bytes of storage by default. Eight bytes of floating point storage provide space for 16 or 17 significant digits. You are not restricted to 8 digits

12 SAS Data Set and Variable Names SAS names can be 32 characters long. can be uppercase, lowercase, or mixedcase. must start with a letter or underscore. Subsequent characters can be letters, underscores, or numeric digits. Valid SAS Names Select the valid default SAS names. data5mon 5monthsdata data#5 five months data fivemonthsdata five_month_data Valid SAS Names Select the valid default SAS names. data5mon 5monthsdata data#5 five months data fivemonthsdata five_month_data SAS Date Values SAS stores date values as numeric values. A SAS date value is stored as the number of days between January 1, 1960, and a specific date. 01JAN JAN JAN1961 store display 01/01/ /01/ /01/ Missing Data Values A value must exist for every variable for each observation. Missing values are valid values. LastName FirstName JobTitle Salary TORRES JAN Pilot LANGKAMM SARAH Mechanic SMITH MICHAEL Mechanic. WAGSCHAL NADJA Pilot TOERMOEN JOCHEN A character missing value is displayed as a blank. A numeric missing value is displayed as a period. 71 Browsing the Data Portion The PRINT procedure displays the data portion of a SAS data set. By default, PROC PRINT displays all observations all variables an Obs column on the left side

13 Browsing the Data Portion General form of the PRINT procedure: PROC PRINT DATA=SAS-data-set; RUN; Example: PROC PRINT Output The SAS System Obs lname baseage gender race weight height BMI 1 Richardson Lowrey Tierney Sommers Kegan proc print data=work.newclass; 73 c02s3d1 74 SAS Data Set Terminology SAS documentation and text in the SAS windowing environment use the following terms interchangeably: SAS Data Set SAS Table SAS Syntax Rules SAS statements usually begin with an identifying keyword always end with a semicolon. DATA newclass; SET mylib.class; Variable Observation Column Row PROC PRINT DATA=newclass; PROC MEANS DATA=newclass; CLASS gender; VAR bmi; 75 RUN; 76 SAS Syntax Rules SAS statements usually begin with an identifying keyword always end with a semicolon. DATA newclass; SET mylib.class; RUN; PROC PRINT DATA=newclass; RUN; PROC MEANS DATA=newclass; CLASS gender; VAR bmi; 77 RUN; SAS Syntax Rules SAS statements are free-format. One or more blanks or special characters can be used to separate words. They can begin and end in any column. A single statement can span multiple lines. Several statements can be on the same line. Unconventional DATA newclass; Spacing SET mylib.class; PROC PRINT DATA=newclass; PROC MEANS DATA=newclass; CLASS gender; VAR bmi; RUN;

14 SAS Syntax Rules Good spacing makes the program easier to read. Conventional Spacing DATA newclass; SET mylib.class; PROC PRINT DATA=newclass; PROC MEANS DATA=newclass; CLASS gender; VAR bmi; SAS Comments Type * to begin a comment. Type your comment text. Type ; to end the comment. * Create work.staff data set and add bmi; DATA newclass; SET mylib.class; * Produce listing report of newclass ; PROC PRINT DATA=newclass; RUN; RUN; Getting Familiar with SAS Data Sets SAS Data Libraries Objectives Explain the concept of a SAS data library. State the difference between a permanent library and a temporary library. Use the CONTENTS procedure to investigate a SAS data library. 83 SAS Data Libraries A SAS data library is a collection of SAS files that are recognized as a unit by SAS. Directory-based A SAS data library is Systems a directory. Windows: c:\mysasfiles UNIX: /users/dept/mysasfiles A SAS data set is a type of SAS file. SAS Data Library SAS File SAS File SAS File 84 14

15 SAS Data Libraries You can think of a SAS data library as a drawer in a filing cabinet and a SAS data set as one of the file folders in the drawer. FILES LIBRARIES Assigning a Libref Regardless of which host operating system you use, you identify SAS data libraries by assigning each a library reference name (libref). libref SAS Data Libraries When you invoke SAS, you automatically have access to a temporary and a permanent SAS data library. work - temporary library sasuser - permanent library ia - permanent library You can create and access your own permanent libraries. work sasuser ia 87 Assigning a Libref You can use the LIBNAME statement to assign a libref to a SAS data library. General form of the LIBNAME statement: LIBNAME libref 'SAS-data-library' <options>; Rules for naming a libref: must be 8 characters or less must begin with a letter or underscore remaining characters are letters, numbers, or underscores. 88 Examples: Assigning a Libref libname ia 'c:\workshop\winsas\prog1'; Making the Connection When you submit the LIBNAME statement, a connection is made between a libref in SAS and the physical location of files on your operating system. Windows 'c:\workshop\winsas\prog1' libname sasclass d:\temp\sasclass'; Libref exists only during the current SAS session

16 Two-level SAS Filenames Every SAS file has a two-level name: libref.filename The data set ia.sales is a SAS file in the ia library. The first name (libref) refers to the library. work Temporary SAS Filename The libref work can be omitted when you refer to a file in the work library. The default libref is work if the libref is omitted. The second name (filename) refers to the file in the library. sales ia sasuser work.employee employee files are deleted when SAS session ends SAS Files SAS data sets and other files are stored in SAS data libraries. SASUSER WORK MYLIB

17 SAS Data Libraries A SAS data library is a collection of SAS files that are recognized as a unit by SAS on your operating environment. WORK - temporary library SASUSER - permanent library You can create and access your own permanent libraries. mylib - permanent library WORK SASUSER MYLIB The LIBNAME Statement LIBNAME statement establishes the library reference (or libref), which is an alias for the SAS data library. global statement - in effect for the entire SAS session until replaced General form of the LIBNAME statement: The libref must be eight characters or fewer. LIBNAME libref 'SAS-data-library' <options>; Ex: LIBNAME mylib your folder ; Two-Level SAS Data Set Names libref.sas-filename Libref SAS Data SAS Data Library SAS DATA SET NAMES 2 level names first level is the libref refers to location of the file Default - file stored in WORK library second level is the member name that identifies the data set within the library Example : work.sales-located in the WORK library 99 mylib.sales-located in the MYLIB library 100 DATA Step PROC Step DATA Step PROC Step SAS Program may consist of a DATA Step Or a PROC Step Or any combination of DATA and PROC Steps EXERCISES II-IV PROC Step

18 DATA Step -- Two Phase Process Phase 1 - Compile phase corresponds to descriptor portion Phase 2 - Execute phase corresponds to data portion 103 Compilation During compilation, SAS checks code for syntax errors translates code to machine code establishes an area of memory called the input buffer if reading raw data establishes an area of memory called the Program Data Vector (PDV) assigns required attributes to variables creates the descriptor portion of the new data set. 104 Execution During the execution phase, SAS initializes the PDV to missing reads data values into the PDV carries out assignment statements and conditional processing writes the observation in the PDV to the output SAS data set at the end of the DATA step (by default) returns to the top of the DATA step initializes any variables that are not read from a SAS data set to missing (by default) 105 repeats the process. The DATA Step Is a Loop Initialize PDV. Execute read statement. Execute program statements. Output observation to SAS data set. End of file? NO YES Stop DATA step. 106 Objectives Reading SAS Data Sets and Creating Variables Create a SAS data set using another SAS data set as input. Create SAS variables. Use operators and SAS functions to manipulate data values. Control which variables are included in a SAS data set

19 Reading a SAS Data Set Create a temporary SAS data set named newclass2 from the permanent SAS data named mylib.class2 and create a variable that represents the bmi. Compute BMI from the variable Height and Weight. mylib.class2 SAS date values New Variable Lname Birthdate Baseage Weight Height BMI Richardson Lowrey Tierney Sommers Reading a SAS Data Set To create a SAS data set using a SAS data set as input, you must use a DATA statement to start a DATA step and name the SAS data set being created (output data set: newclass2) SET statement to identify the SAS data set being read (input data set: myib.class2). To create a variable, you must use an assignment statement to uses the values of the variables Weight and Height and assign the result of the calculation to the variable BMI. 110 Reading a SAS Data Set General form of a DATA step: DATA output-sas-data-set; SET input-sas-data-set; additional SAS statements RUN; By default, the SET statement reads all of the observations from the input SAS data set variables from the input SAS data set. Assignment Statements An assignment statement evaluates an expression assigns the resulting value to a variable. General form of an assignment statement: variable=expression; Define the Variable Name the New Variable new_variable_name = value ASSIGN expression value EVALUATE Rules for naming SAS variables: 1 to 32 characters in length start with a letter (A through Z) or an underscore (_) continue with any combination of numbers, letters, or underscores can be stored in mixed-case

20 An expression contains operands and operators that form a set of instructions that produce a value. Operands are variable names constants. SAS Expressions Operators are symbols that request arithmetic calculations SAS functions. x=3; y=age/10; EXAMPLES assigns 3 to X for all observations assigns the value of age divided by 10 to each observation Clinic= Boston ; assigns the character constant Boston to the variable clinic for each observation bmi= wgtkg/(htm**2); assigns results of calculation to new variable bmi NOTE : X Y and bmi are numeric variables; clinic is a character variable Using Operators Selected operators for basic arithmetic calculations in an assignment statement: Operator Action Example Priority + Addition Sum=x+y; III - Subtraction Diff=x-y; III * Multiplication Mult=x*y; II / Division Divide=x/y; II ** Exponentiation Raise=x**y; I - Negative prefix Negative=-x; I 117 Lname $ 15 Compiling the DATA Step libname mylib 'SAS-data-library'; data newclass2; set mylib.class2; PDV Baseage N 8 Bdate N 8 Race N 8 Weight N 8 Height N PDV lname $ 15 Compiling the DATA Step libname mylib 'SAS-data-library'; data newclass2; set mylib.class2; baseage N 8 Gender N 8 Race N 8 Weight N 8 Height N 8 BMI N Executing the DATA Step Mylib.class2 Lname Birthdate Baseage Weight Height Richardson Lowrey Tierney data newclass2; set mylib.class2; PDV..... newclass

21 Executing the DATA Step PDV Mylib.class2 Lname Birthdate Baseage Weight Height Richardson Lowrey Tierney data newclass2; set mylib.class2; Richardson newclass Executing the DATA Step Mylib.class2 Lname Birthdate Baseage Weight Height Richardson Lowrey Tierney data newclass2; set mylib.class2; PDV Richardson newclass Executing the DATA Step Mylib.class2 Lname Birthdate Baseage Weight Height Richardson Lowrey Tierney Automatic return data newclass2; set mylib.class2; PDV Richardson newclass2 Automatic output Richardson Executing the DATA Step Mylib.class2 Lname Birthdate Baseage Weight Height Richardson Lowrey Tierney. 32 Reinitialize Total data newclass2; to missing set mylib.class2; PDV Richardson newclass2 Richardson Executing the DATA Step Mylib.class2 Lname Birthdate Baseage Weight Height Richardson Lowrey Tierney data newclass2; set mylib.class2; PDV Lowery newclass2 Richardson Executing the DATA Step Mylib.class2 Lname Birthdate Baseage Weight Height Richardson Lowrey Tierney data newclass2; set mylib.class2; PDV Lowery newclass2 Richardson

22 Executing the DATA Step Mylib.class2 Lname Birthdate Baseage Weight Height Richardso Lowrey Tierney Automatic return data newclass2; set mylib.class2; PDV Lowery newclass2 Automatic output Richardson Lowery Assignment Statements proc print data=newclass2; format birthdate date9.; var lname birthdate height weight bmi; Obs lname birthdate height weight BMI 1 Richardson 03FEB Lowrey 23MAR Tierney Sommers 10JUN Kegan 08AUG Why is BMI missing in observation 5? 128 Printing Dates PROC PRINT data=example; Var id birthdate; Format mmddyy10.; SPECIFY FORMAT Run; Obs id birthdate /04/ /28/1991 Without format SAMPLE SAS DATA FORMATS Print: SAS format: MMDDYY6. 04/01/97 MMDDYY8. 04/01/1997 MMDDYY DDMMYY6. 01/04/97 DDMMYY8. 01/04/1997 DDMMYY10. 01APR97 DATE7. 01APR1997 DATE Portions Copyright 2001 SAS Institute Inc., Cary, NC, USA. All Rights Reserved. Reproduced with permission from SAS Institute Inc., Cary NC, USA. 22