new in sas 9.2

Upload: rajesh-kumar

Post on 14-Apr-2018

232 views

Category:

Documents


0 download

TRANSCRIPT

  • 7/29/2019 New in SAS 9.2

    1/33

    NEW IN SAS 9.2

    STATISTICAL PROCEDURES

    SUN LI

    SENIOR STATISTICIANCENTRE FOR ACADEMIC COMPUTING

    [email protected]

    mailto:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]
  • 7/29/2019 New in SAS 9.2

    2/33

    OUTLINE

    Useful links

    Base SAS

    SAS/ACCESS

    SAS/STAT

    SAS Stat Studio software

    SAS Power and Sample Size (PSS application)

    PROC GENMOD Bayesian capability

    PROC GLIMMIX Generalized linear mixed model

    SAS/ETS : PROC COUNTREG - ZINB model

    SAS/OR: SAS Simulation Studio2

  • 7/29/2019 New in SAS 9.2

    3/33

    USEFUL LINKS

    SAS 9.2 Documentation:

    Whats new in SAS 9.2?

    http://support.sas.com/documentation/cdl/en/whatsnew/61982/HTML/default/acpcrefwhatsnew902.htm

    http://support.sas.com/documentation/cdl/en/whatsnew/62435/PDF/default/whatsnew.pdf

    Training notes & example data sets:http://research2.smu.edu.sg/CAC/StatisticalComputing/Wiki/SAS-Training%20Slides%20and%20Syntax.aspx

    3

    http://support.sas.com/documentation/cdl/en/whatsnew/61982/HTML/default/acpcrefwhatsnew902.htmhttp://support.sas.com/documentation/cdl/en/whatsnew/61982/HTML/default/acpcrefwhatsnew902.htmhttp://support.sas.com/documentation/cdl/en/whatsnew/62435/PDF/default/whatsnew.pdfhttp://support.sas.com/documentation/cdl/en/whatsnew/62435/PDF/default/whatsnew.pdfhttp://research2.smu.edu.sg/CAC/StatisticalComputing/Wiki/SAS-Training%20Slides%20and%20Syntax.aspxhttp://research2.smu.edu.sg/CAC/StatisticalComputing/Wiki/SAS-Training%20Slides%20and%20Syntax.aspxhttp://research2.smu.edu.sg/CAC/StatisticalComputing/Wiki/SAS-Training%20Slides%20and%20Syntax.aspxhttp://research2.smu.edu.sg/CAC/StatisticalComputing/Wiki/SAS-Training%20Slides%20and%20Syntax.aspxhttp://research2.smu.edu.sg/CAC/StatisticalComputing/Wiki/SAS-Training%20Slides%20and%20Syntax.aspxhttp://research2.smu.edu.sg/CAC/StatisticalComputing/Wiki/SAS-Training%20Slides%20and%20Syntax.aspxhttp://support.sas.com/documentation/cdl/en/whatsnew/62435/PDF/default/whatsnew.pdfhttp://support.sas.com/documentation/cdl/en/whatsnew/62435/PDF/default/whatsnew.pdfhttp://support.sas.com/documentation/cdl/en/whatsnew/61982/HTML/default/acpcrefwhatsnew902.htmhttp://support.sas.com/documentation/cdl/en/whatsnew/61982/HTML/default/acpcrefwhatsnew902.htm
  • 7/29/2019 New in SAS 9.2

    4/33

    Enhanced procedures to discuss:

    PROC MEANS

    PROC UNIVARIATE

    PROC FREQ

    4

  • 7/29/2019 New in SAS 9.2

    5/33

    BASE SAS

    1. PROC MEANS

    The PRT statistic is now an alias for the PROBT statistic

    The MODE statistic can now be used with PROC MEANS.

    5

  • 7/29/2019 New in SAS 9.2

    6/33

    BASE SAS

    2. PROC UNIVARIATE

    CDFPLOT: plots the observed cdf of a variable and enables you tosuperimpose a fitted theoretical distribution on the graph.

    PPPLOT: creates a P-P plot which compares the ecdf of a variable

    with a specified theoretical cumulative distribution function.

    6

  • 7/29/2019 New in SAS 9.2

    7/33

    BASE SAS

    3. PROC FREQ

    Testing for specified proportions

    Distribution plot and other plots

    Binomial proportion tests and confidence intervals

    7

  • 7/29/2019 New in SAS 9.2

    8/33

    Support new file formats:

    dBase with memos (DBFMEMO), JMP, Paradox DB, SPSS

    SAV, Stata DTA

    Date/Time Value:

    USEDATE=YES /NO option in PROC IMPORT

    8

  • 7/29/2019 New in SAS 9.2

    9/33

    New related software:

    SAS Stat Studio software

    SAS Power and Sample Size (PSS application)

    New procedures:

    Bayesian capabilities are introduced under procedures:GENMOD, LIFEREG, and PHREG

    GLIMMIX, GLMSELECT, and QUANTREG

    Experimental procedures: HPMIXED, MCMC, SEQDESIGN, SEQTEST ,TCALIS

    9

  • 7/29/2019 New in SAS 9.2

    10/33

    SAS/STAT

    1. SAS Stat Studio 3.1

    New software for data exploration and analysis

    Start > All Programs > Stat Studio 3.1

    Demo basic steps with hands-on: Graph & Polynomial regression with data set:

    hurricanes.sas7bdat

    Transform data & Model fitting with data setbaseball.sas7bdat

    SAS help and documentation:

    SAS products -> SAS Stat Studio 3.1 -> SAS Stat Studio 3.1Users Guide

    10

  • 7/29/2019 New in SAS 9.2

    11/33

    11

  • 7/29/2019 New in SAS 9.2

    12/33

    SAS/STAT

    2. SAS Power and Sample Size 3.1

    PSS application

    Start > All Programs > SAS > SAS Power and Sample Size> SAS Power and Sample Size 3.1

    Demo basic steps with hands-on:

    Power analysis for One Sample T Test

    SAS help and documentation:

    SAS products -> SAS/STAT -> SAS/STAT Users Guide ->The Power and Sample Size Application

    12

  • 7/29/2019 New in SAS 9.2

    13/33

    13

  • 7/29/2019 New in SAS 9.2

    14/33

    SAS/STAT

    3. PROC GENMOD Bayesian Analysis of a Linear Regression Model:

    Here is a study of 54 patients undergoing a certain kind of liveroperation in a surgical unit. The data setSurg contains survival

    time and certain covariates for each patient.

    14

  • 7/29/2019 New in SAS 9.2

    15/33

    SAS/STAT

    15

  • 7/29/2019 New in SAS 9.2

    16/33

    SAS/STAT

    16

    BAYES : produces Bayesian analysis via Gibbs sampling for most of thestatistical analyses.

    SEED: specifies randomization seed. It is used to maintain reproducibility.

    OUTPUT : saves the samples in the SAS data set PostSurg for furtherprocessing.

    By default, a uniform prior distribution is assumed on the regressioncoefficients. A noninformative gamma prior is used for the scale parametersigma.

    ODS Graphics is enabled as specified in the SAS statements to display thediagnostic plots.

  • 7/29/2019 New in SAS 9.2

    17/33

  • 7/29/2019 New in SAS 9.2

    18/33

    18

    SAS/STAT

    There is a 1.00 probability of a positive

    relationship between the logarithm of a

    blood clotting score and survival time,adjusted for the other covariates.

  • 7/29/2019 New in SAS 9.2

    19/33

    19

    SAS/STAT

    4. PROC GLIMMIX

    Comparison btw GLIMMIX and other procedures

    1) Response can have a nonnormal distribution (MIXED assumesnormally distributed response.)

    2) Incorporates random effects in the model and so allows forconditional and marginal inference (GENMOD allows only formarginal inference.)

    3) The class of generalized linear mixed models is a special case of the

    nonlinear mixed models; hence some of the models you can fit withNLMIXED can also be fit with GLIMMIX.

    (The details can be found in the SAS help documentation.)

  • 7/29/2019 New in SAS 9.2

    20/33

    Logistic Regressions with Random Intercepts

    Researchers investigated the performance of banks in a multicenterstudy. They randomly selected 15 centers. One of the study goals wasto compare the occurrence of loan defaults for the banks. In eachcenter two types of banks were selected and coded as A" and "B.Under each type, there are a few of banks selected for the study.

    20

    SAS/STAT

  • 7/29/2019 New in SAS 9.2

    21/33

    To model the probability of defaults happened in the two types of banks:A andB, you need to account for the fixed group effect and the randomselection of centers. We assume a linear model that relates group andcenter effects to the logit of the probabilities:

    A-B measures the difference in the logits.

    0 is the overall intercept in the model. 21

    SAS/STAT

  • 7/29/2019 New in SAS 9.2

    22/33

    22

    SAS/STAT

    CLASS : specifies the classification variables.

    MODEL : specifies response variable as a sample proportion by using theevents/n syntax.

    SOLUTION : requests a list of solutions for fix-effects parameter estimates.Because the default/n syntax, the procedure gives binomial distribution withlogit link.

    RANDOM : specifies that the linear predictor contains an intercept term that

    randomly varies at the level ofcentereffect. In other words, a randomintercept is drawn separately and independently for each center in the study.

    LSMEANS : requests the least squares means of the group effect on the logitscale.

    ILINK: adds estimates, standard errors, and confidence limits on the mean(probability) scale .

  • 7/29/2019 New in SAS 9.2

    23/33

    23

    SAS/STAT

  • 7/29/2019 New in SAS 9.2

    24/33

    24

    SAS/STAT

    The details can be found in the SAS help documentation in the

    individual procedures.

  • 7/29/2019 New in SAS 9.2

    25/33

    25

  • 7/29/2019 New in SAS 9.2

    26/33

  • 7/29/2019 New in SAS 9.2

    27/33

    27

    SAS/ETS

    ZIP and ZINB Models for Data Exhibiting Extra Zeros

    An often encountered characteristic of count data is that the number ofzeros in the sample exceeds the number of zeros predicted by either thePoisson or negative binomial model.

    Zero-inflated Poisson (ZIP) and zero-inflated negative binomial (ZINB)models explicitly model the production of zero counts to account forexcess zeros.

  • 7/29/2019 New in SAS 9.2

    28/33

    28

    SAS/ETS

    This study examines how factors such as gender (fem), marital status (mar),number of young children (kid5), prestige of the graduate program (phd),and number of articles published by a scientists mentor (ment), affect thenumber of articles (art) published by the scientist.

  • 7/29/2019 New in SAS 9.2

    29/33

    29

    SAS/ETS

  • 7/29/2019 New in SAS 9.2

    30/33

    30

    SAS/ETS

  • 7/29/2019 New in SAS 9.2

    31/33

    31

    SAS Simulation Studio

    Other enhanced features, see theSAS/OR documentation.

  • 7/29/2019 New in SAS 9.2

    32/33

    32

  • 7/29/2019 New in SAS 9.2

    33/33