SAS 9'0 - PowerPoint PPT Presentation

1 / 30
About This Presentation
Title:

SAS 9'0

Description:

var Exponential_x; histogram /exp(fill l=3) cfill=yellow midpoints=.05 to 5.55. by .25 ... slope var = 0 noest; season length = 12 type=trig; estimate back=24; ... – PowerPoint PPT presentation

Number of Views:106
Avg rating:3.0/5.0
Slides: 31
Provided by: HALL2
Category:
Tags: sas | var

less

Transcript and Presenter's Notes

Title: SAS 9'0


1
SAS 9.0
  • April 9, 2003
  • Linda Atkinson

2
Documentation
  • Online Documentation combined with Help
  • Quick-Start Guide V9 Getting Started with SAS
  • Resource Guide for experienced SAS programmers
  • Documentation on web http//v9doc.sas.com/sasdoc
  • Base SAS Community
  • http//support.sas.com/rnd/base/index-v9-resources
    .html
  • Papers by SAS developers
  • The DATA step in Version 9 Whats New?, Jason
    Secosky
  • V9 Nuts Bolts, David Shamlin
  • A Preview of SAS/STAT Version 9, Maura Stokes
    et al

3
Highlights
  • SAS now runs in many 64-bit operating
    environments
  • Parallel processing. Threaded procedures in V9
  • PROC REG
  • PROC GLM
  • PROC LOESS
  • PROC DMREG
  • PROC DMINE
  • PROC SORT
  • PROC SUMMARY/MEANS

4
Highlights (cont.)
  • New SAS/GRAPH styles provide a consistent look
    for output created by the Output Delivery System
    (ODS).
  • The XML engine imports a wider variety of
  • XML documents.
  • The Output Delivery System writes to more
  • destinations and provides more formatting
    selections.

5
Highlights (cont.)
  • Format and informat names can be more
  • than eight characters
  • New functions
  • New analytic tasks in Enterprise Guide, more
    controls, new graph styles, extensibility features

6
Output Delivery System
  • DOCUMENT destination PROC DOCUMENT
  • New styles
  • MARKUP destination
  • ODS statistical graphics NOT YET

7
PROC DOCUMENT Example
  • options nodate nonumber
  • data distrdata
  • drop n
  • label Normal_x'Normal Random Variable'
  • Exponential_x'Exponential Random
    Variable'
  • do n1 to 100
  • Normal_x10rannor(53124)50
  • Exponential_xranexp(18746363)
  • output
  • end
  • run
  • ods document nameuniv(write)
  • title '100 Obs Sampled from a Normal
    Distribution'

8
PROC DOCUMENT (Cont.)
  • proc univariate datadistrdata noprint
  • var Normal_x
  • histogram Normal_x/normal(noprint)
    cbarlinegrey
  • name"normal"
  • run
  • title '100 Obs Sampled from an Exponential
    Distribution'
  • proc univariate datadistrdata noprint
  • var Exponential_x
  • histogram /exp(fill l3) cfillyellow
    midpoints.05 to 5.55
  • by .25 name"exp"
  • run
  • ods document close

9
PROC DOCUMENT (Cont.)
  • title
  • proc document
  • doc nameuniv
  • list/levelsall
  • run
  • quit
  • proc document nameuniv
  • ods pdf file'd\sas 9 demo\univ.pdf'
  • replay \univariate1\Normal_x1\Normal1
  • run
  • quit
  • ods pdf close

10
New ODS Styles
11
Page of option for RTF output
  • options nodate nonumber


  • ods escapechar''


  • ods rtf file'd\sas 9 demo\file.rtf'

  • title " Sales Data" jr "Page pageof"

  • proc print datasashelp.prdsale(obs200)

  • run


  • ods rtf close




12
Excel spreadsheets
  • EXPORT procedure can specify the SHEET option to
    specify a specific spreadsheet in a workbook
  • Engine on LIBNAME statement allows direct access
    to Excel data

13
XML support
  • New, more capable LIBNAME engine SXLE imports a
    broader variety of XML documents
  • XML Atlas (beta) is a Java application that
    assists you in creating and modifying XMLMap files

14
New mathematical functions
  • BETA returns the value of the beta function
  • LOGBETA returns the logarithm of the beta
    function
  • IQR returns the interquartile range
  • LARGEST returns the kth largest nonmissing value
  • MAD returns the median absolute deviation from
    the median

15
Mathematical functions (cont.)
  • MEDIAN computes median values
  • PCTL computes percentiles
  • SMALLEST returns the kth smallest nonmissing value

16
STAT Enhancements
  • Multiple imputation procedures are production,
    include a TEST statement to test linear
    hypotheses about the parameters and an
    experimental CLASS statement to specify
    categorical variables
  • Experimental ROBUSTREG procedure to analyze data
    that may include outliers
  • TPHREG test release of PHREG incorporating the
    class statement

17
STAT Procedure Enhancements
  • New STRATA statement in LOGISTIC procedure for
    conditional logistic regression, scoring of data
    sets through new SCORE statement
  • GLM ability to form classification groups using
    the full formatted length of the CLASS variable
    levels
  • New SURVIVAL statement in the LIFETEST procedure

18
STAT (other new procedures)
  • DISTANCE procedure (experimental) to compute
    measures of distance, similarity or dissimilarity
    between the observations of a SAS data set
  • POWER procedure experimental to perform power and
    sample size computations.

19
ROBUSTREG example
  • proc robustreg datastack
  • model y x1 x2 x3 / diagnostics leverage
  • id x1
  • test x3
  • run

20
DISTANCE example
  • proc distance datastock methoddcorr
    outdistdcorr
  • var interval(div_1986 div_1987 div_1988
    div_1989 div_1990)
  • id compname
  • run

21
ETS New Procedures
  • ENTROPY provides Generalized Maximum Entropy
    estimations for linear systems of equations
  • TIMESERIES procedure analyzes time-stamped
    transactional data with respect to time
  • UCM provides estimation for Unobserved Component
    Models (Structural Models)

22
ENTROPY example
  • proc entropy datarate
  • model ya b c d ab ac ad bc bd cd abc abd acd
    bcd abcd
  • run

23
TIMESERIES Example
  • proc timeseries datasashelp.air outseries
  • outtrendtrend outseasonseason
    printseasons
  • id date intervalqtr accumulateavg
  • var air
  • run

24
UCM Example

data series_g set sashelp.air logair
log(air) run proc ucm data series_g id date
interval month model logair irregular
level slope var 0 noest season length 12
typetrig estimate back24 forecast back12
lead24 printforecasts run
25
Other ETS enhancements
  • SASEHAVR interface engine for accessing data in a
    Haver Analytics Data Link Express (DLX) data base
  • New functions (financial and date/time)
  • OUTLIER statement in PROC ARIMA
  • New transformation operators in EXPAND
  • MDC has new syntax on RESTRICT statement, new
    BOUNDS statement
  • MODEL has SMM (Simulated Method of Moments)
    estimation

26
ETS enhancements (Cont.)
  • New features in VARMAX
  • X12 procedure default behavior has changed with
    regard to missing leading and trailing values
    new statements and options
  • Time Series Forecasting System now supports
    factored ARIMA models

27
SAS/GRAPH enhancements
  • New GBARLINE procedure to create vertical bar
    chars with line plot overlays

28
SAS/GRAPH enhancements
  • New GAREABAR procedure generates bar graphs in
    which the width of the bars are proportional to
    some characteristic of the data element
    represented by the bars

29
SAS/GRAPH enhancements
  • New options in PIE statement in GCHART enables
    you to create detail pie charts, which have an
    inner pie overlay

30
Another SAS V9 presentation
  • http//support.sas.com/videos/sas_v9
  • SAS System 9 The Revolution Continues
  • By Jim Metcalf, Director, Foundation Technology
    Strategy, SAS Worldwide Marketing
Write a Comment
User Comments (0)
About PowerShow.com