ppt - PowerPoint Presentation

Download Report

Transcript ppt - PowerPoint Presentation

Efficient Way to Learn SAS
with Virtually No Cost
Mirjana Stojanovic
Statistical Programmer
Chapel Hill, NC
M Stojanovic:
M Stojanovic:
Learning from WEB sites
 Find
out WEB sites which is the
most suitable for your needs
 Follow instructions or copy and
paste into SAS learning edition
(optional)
 Check back frequently and check
links which you will use the most of
the time
FirstObs.com
http://www.firstobs.com
 Detailed
and compact
 Broad spectrum of information
 Good links to other sites
 Not too narrative
 Basic and advanced procedures
FirstObs.com (cont.)
http://www.firstobs.com

Tutorial List
 Net Resources expanded list
of Internet resources
 SAS Community
 Search FirstObs.com site by
keyword and topic
FirstObs.com (cont.)
http://www.firstobs.com
Data step
 Combining data files
 Execute: if...Then
 Functions
 Reading raw data
 Retain statement
FirstObs.com (cont.)
http://www.firstobs.com
SAS procedure
Proc Format
Proc Freq
Proc Print
Proc Sort
Proc Contents
Institute for Clinical and Epidemiologic Research
http://hsrd.durham.med.va.gov/Biostat/training/SAS%20presentation%20v10.pdf

Comprehensive manual
 Easy to follow
 Simple sentences
 Descriptive explanation
 Programming mistakes and how
to avoid them
Institute for Clinical and Epidemiologic Research
http://hsrd.durham.med.va.gov/Biostat/training/SAS%20presentation%20v10.pdf
Basic Components of SAS
- Getting Data into a SAS Data Set
- Types of SAS Data Sets
- Permanent Data Sets - Libname
Statement
- Using Existing SAS Data Set
* Set Statement
* Merge Statement
Institute for Clinical and Epidemiologic Research
http://hsrd.durham.med.va.gov/Biostat/training/presentation%20II%20v10.pdf
Introduction to SAS Procedures
 Proc Means
 Proc Univariate
 Proc Freq
 Proc Contents
 Proc Plot
Virginia Commonwealth University
http://www.people.vcu.edu/~rjohnson/bios524/Cody%20Notes/index.html
 Based
on Cody and Pass, SAS
Programming by Example
 Very clear and simple sentences
 Descriptive explanation
 Easy navigation
 Remarks/Hints in red color
 One example per page
Virginia Commonwealth University (cont.)
http://www.people.vcu.edu/~rjohnson/bios524/Cody%20Notes/index.html
Chapter 1.
What goes on in the data step?
- The compilation phase
- The execution phase
Chapter 8.
“Remembering" values from
previous observations
RHO, Inc.
http://www.rhoworld.com/index_services_biostats.htm
Explanation through LOG messages(Note)
Miscellaneous programming materials
Tutorial on the SAS system
Good summary at the end of chapters
Examples supported with explanations
Chapter 1.
Introduction
Chapter 2.
Introduction to Computers
Chapter 3.
Introduction to SAS
Chapter 4.
Using SAS Procedures
Chapter 5. Transforming SAS
Chapter 6.
Data Sets
SAS Programming
Chapter 7.
Managing and Combining SAS Data Sets
Chapter 8.
Using External Files with SAS
Chapter 9.
Report Writing
Chapter 10.
Matrix Algebra with SAS/IML
Chapter 11.
Statistical Graphics with SAS/Graph
Chapter 12.
Data Entry and Editing with SAS/FSP
Chapter 13. The
Macro Language
Chapter 14. Miscellaneous Programming
Chapter 15.
ODS
Research Data Management
PAPER
Materials
RHO, Inc.
http://www.rhoworld.com/index_services_biostats.htm
Chapter 5.
 Transforming SAS Data Sets
a. Creating new SAS data sets
b. Creating and transforming
variables
c. Subsetting observations
d. Subsetting variables
WHERE vs IF
The WHERE
statement works before observation are brought into the
data step(that is the PROGRAM DATA VECTOR) .
The
IF statement works on observation that are already in the data step.
The WHERE
statement is not executable, but the IF statement is
The WHERE
statement operates only on observations in SAS data sets,
whereas the IF statement can operate either on observations from existing
SAS data sets or on observations created with an input statement.
If
a BY statement does not accompany a SET or MERGE statement, the
WHERE and IF statements usually produce the same result
In
almost all cases a WHERE statement is more efficient than an IF
statement(observations do not have to be moved into the PDV)
The WHERE
PROCS .
statement, but not the IF statement can be used in SAS
RHO, Inc.
http://www.rhoworld.com/index_services_biostats.htm
Chapter 7.
COMBINING AND MANAGING
SAS DATA SETS
a. Multiple input data sets
b. SAS special variables
c. Reshaping SAS data sets
d. Data library management
RHO, Inc.
http://www.rhoworld.com/index_services_biostats.htm
MISCELLANEOUS PROGRAMMING
MATERIALS
University of Central Florida
http://dms.stat.ucf.edu/Sta4102notes/Prog2/Chapter10/STA4102-Chapter10.ppt
Chapter 10
 Well animated
 Step by step explanations
 Refresher Quiz
 Different color for better
understanding
University of Central Florida(cont.)
http://dms.stat.ucf.edu/Sta4102notes/Prog1/Chapter5/Chapter5-STA4102.ppt
 Chapter
5
Process flow of Subseting IF
How to create SAS variable conditionally
University of Central Florida
(cont.)
http://dms.stat.ucf.edu/Sta4102notes/Prog2/Chapter9/Chapter9-STA4102.ppt
 Chapter
9
Refresher Quiz
Karolinska Institutet in Stockholm.
Paul Dickman
 http://www.pauldickman.com/teaching/sas/i
ndex.html
 http://www.pauldickman.com/teaching/sas/s
et_by.html
 http://www.pauldickman.com/teaching/sas/s
as_tips_cleaning8.pdf
 http://www.pauldickman.com/teaching/sas/s
as_tips_cleaning8.pdf
National Institute of Health
http://support.cit.nih.gov/statsoft/training/classrm/courstmp.htm
SAS Programming Fundamentals I
http://support.cit.nih.gov/Statsoft/training/crsn
otes/pdf/sasfund1.pdf
SAS Programming Fundamentals II
http://support.cit.nih.gov/statsoft/training/crsnote
s/fun22tmp.htm
National Institute of Health (cont.)
 Systematic
explanations
 Plenty of good examples
 Exercises included
 Creating maps with SAS
National Institute of Health
http://support.cit.nih.gov/statsoft/training/classrm/courstmp.htm
Producing Tables with SAS
Producing Graphs with SAS
Producing Reports with SAS
Using the SAS System for Elementary
Statistical Analysis
Using Proc Freq in the SAS System to
Perform Categorical Data Analysis
University of Florida
http://ifasstat.ufl.edu/sta4702/pdf/SASIntro.pdf
 Descriptive
 Choosing
different color for a better
understanding
 Presentation in pictures
University of Florida (cont.)
http://ifasstat.ufl.edu/sta4702/pdf/SASIntro.pdf
 Rules
of SAS Use
 General Form of a SAS Program
 Different ways one can read in data
using SAS
 Reading Data from an External Text
File
 SAS Datasets
University of Florida (cont.)
http://ifasstat.ufl.edu/sta4702/pdf/SASIntro.pdf
 Temporary
and Permanent SAS Data
Sets
 Creating a Permanent Data Set
 List Input
 Delimited Files
 Combining Data Sets
Andrew Karp on Proc Means
 http://www.hasug.org/HASUG%2
0_PROC_MEANS_Enhancements
.pdf
 http://www2.sas.com/proceeding
s/sugi27/p018-27.pdf
 http://www.dasconsultants.com/
papers/CarpMcQProcMeans.pdf
SAS Customer Support Center
 http://support.sas.com/documentation/onlin
edoc/code.samples.html
 http://support.sas.com/publishing/pdf/57198
_ch3pg76.pdf
Using PROC MEANS and PROC FREQ
to count MISSING VALUES
The Pennsylvania State University
http://www.stat.psu.edu/~jianchu/STAT480.htm#lecture%20Notes
 Clear
short examples
 Homework's and solutions
 Examples on one page
 Easy to follow
 Given advantages
The Pennsylvania State University (cont.)
http://www.stat.psu.edu/~jianchu/STAT480.htm#lecture%20Notes
http://www.stat.psu.edu/~jianchu/Get
tingStartted_files/frame.htm
Getting data into the SAS system
Working with your SAS data: Part I
Working with your SAS data: Part II
Working with multiple SAS data sets
Working with statistics
The Pennsylvania State University
http://www.stat.psu.edu/~dcsteven/stat480/
Quiz 1 (with correct answers)
Quiz 2 (with correct answers)
http://www.sasusers.com/links/default
 Information & Resources
for SAS® Users
 Stats & SAS Links:
 General Topics
http://www.sys-seminar.com/SASArticles.htm
Newsletter “Missing semicolon”
SAS Tips
Small but useful examples, SAS news
Keep your attention from the beginning to
the end
SAS: the SAS Users'
Group List (SUG-L)
What is SUG-L?
SUG-L is a listserv-based mailing list for SAS
users throughout the Research Triangle area.
The list provides a way for users in the area to
exchange information about SAS problems and
questions.
 Meetings of The Research Triangle SAS
Users Group (RTSUG) are announced on
this list.
SAS Newsgroup
comp.soft-sys.sas
This newsgroup and SAS-L are
joined(mirrored) so
that the messages sent to one will also be
sent to the other.
Deja.com Power Search provides an easy
way to read comp.soft-sys.sas and search
the archives.
Philip Mason
http://homepage.ntlworld.com/philipmason/
 SAS Tips
and Techniques
 SAS related links
 SAS Papers
Conclusion



WEB is a golden mine for a learning
WEB plus combination with other
resources will give a best results
Complementary ways:
- BBU books
- SUGI, SESUG, PHARMASUG
- SAS Training classes
- Attending SAS Local groups
- Actively using SUG-L
- WORKING EXPERIENCE
Mirjana Stojanovic
Statistical programmer
Chapel Hill, NC 27514
[email protected]
Tel: (919) 932-6236