超经典的SAS BASE的笔记3Word格式文档下载.docx
《超经典的SAS BASE的笔记3Word格式文档下载.docx》由会员分享，可在线阅读，更多相关《超经典的SAS BASE的笔记3Word格式文档下载.docx（41页珍藏版）》请在冰豆网上搜索。
;
dataif_01_b;
lengthtype$8.;
setif_01_a;
ifcode=1thentype='
Fix'
ifcode=2thentype='
Variable'
ifcode^=1andcode^=2thentype='
Unknown'
labeltype='
TypesofMortgageRate'
run;
IF-THEN/ELSE*/
dataif_01_c;
elseifcode=2thentype='
elseifcode^=1andcode^=2thentype='
Letusseehowtodivideagegroup?
dataage_grp;
inputpat_idage@@;
2901566629987168280256642704566026215658
2632565526645653250656442512564325745647
2583564824460642249456412332563323765637
228356282226062221985621
dataage_grp2;
lengthagegrp$5.;
setage_grp;
if20<
=age<
35thenagegrp='
20-34'
elseif35<
50thenagegrp='
35-49'
elseif50<
65thenagegrp='
50-64'
elseif65<
=agethenagegrp='
65+'
procfreq;
tablesagegrp/list;
*tablesagegrp/out=age_grp3;
Advice:
AbetterwaytowritemultipleIFstatementistouseanELSEbeforeallbutthefirstIF.TheotherIF-THEN/ELSEstatementswouldlooklikethis:
Ifthen;
Elseifthen;
…………..;
TheeffectoftheELSEstatementsisthatwhenanyIFstatementistrue,allthefollowingELSEstatementsareskipped.Theadvantageistoreducecomputertime（sincealltheIFdonothavetobetested）andevenlytoavoidthefollowingtypeoferror.Wouldyouseewhatwillhappenwiththestatementsbelow?
dataer;
inputx@@;
cards;
12345
dataer_2;
seter;
ifx=1thenx=5;
ifx=2thenx=4;
ifx=4thenx=2;
ifx=5thenx=1;
dataer_3;
elseifx=2thenx=4;
elseifx=4thenx=2;
elseifx=5thenx=1;
IF-THEN/DO*/
dataoss_test;
lengthcourse$8.school$4.;
inputcourseschoolscoreyear;
EnglishNT921998
EnglishNT941999
EnglishNT962000
EnglishNT912001
EnglishNSS881998
EnglishNSS801999
EnglishNSS842000
EnglishNSS822001
MathNT861998
MathNT881999
MathNT892000
MathNT862001
MathNT902002
MathNSS841998
MathNSS881999
MathNSS882000
MathNSS922001
MathNSS892002
dataoss_test1;
setoss_test;
ifcourse='
English'
thendo;
ifscore>
84thenoutput;
end;
dataoss_test2;
ifschool='
NT'
ifcourse='
Math'
ifscore>
86thenoutput;
EarlieryoulearnedtoassignvaluesconditionallyusingIF-THEN/ELSEstatements.YoucanalsouseSELECTgroupsinDATAstepstoperformconditionalprocessing.ASELECTgroupcontainsthesestatements:
Thisstatement...
Performsthisaction...
SELECT
beginsaSELECTgroup.
WHEN
identifiesSASstatementsthatareexecutedwhenaparticularconditionistrue.
OTHERWISE（optional）
specifiesastatementtobeexecutedifnoWHENconditionismet.
END
endsaSELECTgroup.
SELECT<
（select-expression）>
WHEN-1（when-expression-1<
...,when-expression-n>
）statement;
WHEN-n（when-expression-1<
OTHERWISEstatement;
END;
Example:
datasel;
inputidsalarygenderjob$;
128000CA
231000RN
326981ME
445501MD
538951TA
datasel2;
lengthoccupation$20Sex$7;
setsel;
select（id）;
when
（1）income=salary*10;
when（3,4）income=salary*15;
otherwiseincome=salary*5;
select（job）;
when（'
CA'
）occupation="
CharteredAccountant"
RN'
RegistedNurse"
ME'
MechanicI"
MD'
Doctor"
otherwiseoccupation="
Other"
select（gender）;
when（0）Sex="
Female"
when
（1）Sex="
Male"
otherwiseSex='
BasicConcept:
2.1.1SortingOrdersforNumericVariables
Fornumericvariables,thesmallest-to-largestcomparisonsequenceis
1.SASSystemmissingvalues（shownasaperiodorspecialmissingvalue）
2.negativenumericvalues
3.zero
4.positivenumericvalues.
datasorting_1;
.10-23
procsortdata=sorting_1;
byx;
orderofoutput:
.-2013;
2.1.2SortingOrdersforCharacterVariables
datasorting_2;
inputarea$@@;
torontoLondon535.hamilton
procsortdata=sorting_2;
byarea;
Output:
‘blank’,535,London,hamilton,toronto
2.2WhatcanSORTdo?
-Specifytheinputdataset
-Createanoutputdataset
-Specifytheoutputorder
-EliminateduplicateobservationswithcommonBYvaluesandotheroptions
2.3ApplicationsofPROCSORTprocedure
ThesortproceduresortsobservationsinaSASdatasetbyoneormorecharacterornumericvariables,eitherreplacingtheoriginaldatasetorcreatinganew,sorteddataset.PROCSORTbyitselfproducesnoprintedoutput.
2.3.1ObservationsSortedbytheValuesofOneVariable
Inthisexample,PROCSORTreplacestheoriginaldataset,sortedalphabeticallybylastname,withadatasetthatissortedbyemployeeidentificationnumber.Thestatementsthatproducetheoutputfollow:
datasorting_3;
inputName$IDnumber;
datalines;
Arnsbarger5466
Belloit1988
Capshaw7338
Lemeux4210
Pierce5779
Wesley2092
procsort;
byidnumber;
2.3.2ObservationsSortedbytheValuesofMultipleVariables
Thebusinessesinthisexamplearefirstsortedbytown,thenbydebtfromhighestamounttolowestamount,thenbyaccountnumber.
DESCENDINGoption:
reversesthesortorderforthevariablethatimmediatelyfollowsinthestatementsothatobservationsaresortedfromthelargestvaluetothesmallestvalue.
datasorting_4;
inputcompany$1-23town$24-36debtaccnt_num;
datalines;
ApexCateringApex37.959923
Bob'
sBedsMorrisville119.954998
Boyd&
SonsAccountingGarner312.494762
DeluxeHardwareGarner467.128941
ElwayPianoandOrganGarner65.795217
IceCreamDelightHollySprings299.982310
Pauline'
sAntiquesMorrisville302.059112
Paul'
sPizzaApex83.001019
Peter'
sAutoPartsApex65.797288
StricklandIndustriesMorrisville657.221675
Tina'
sPetShopApex37.955108
Tim'
sBurgerStandHollySprings119.956335
WatsonTaborTravelApex37.953131
WorldWideElectronicsGarner119.951122
bytowndescendingdebt;
2.3.3CreateOutputDataSetfortheSortedObservations
procsortdata=sorting_4out=sorting_5;
2.3.4Eliminateduplicateobservations
-NODUPKEYoption
Inthisexample,PROCSORTcreatesanoutputdatasetthatcontainsonlythefirstobservationofeachBYgroup.TheNODUPKEYoptionremovesanobservationfromtheoutputdatasetwhenitsBYvalueisidenticaltothepreviousobservation'
sBYvalue.Theresultingreportcontainsoneobservationforeachtownwherethebusinessesarelocated.ItautomaticallyeliminatesmultipleobservationswheretheByvariableshavethesamevalue.
optionsnodatepageno=1linesize=80pagesize=60;
dataaccount;
inputCompany$1-22Debt25-30AccountNumber33-36
Town$39-51;
sPizza83.001019Apex
WorldWideElectronics119.951122Garner
StricklandIndustries657.221675Morrisville
IceCreamDelight299.982310HollySprings
WatsonTaborTravel37.953131Apex
SonsAccounting312.494762Garner
sBeds119.954998Morrisville
sPetShop37.955108Apex
ElwayPianoandOrgan65.795217Garner
sBurgerStand119.956335HollySprings
sAutoParts65.797288Apex
DeluxeHardware467.128941Garner
sAntiques302.059112Morrisville
ApexCatering37.959923Apex
procsortdata=accountout=townsnodupkey;
bytown;
procprintdata=towns;
vartowncompanydebtaccountnumber;
title'
TownsofCustomerswithPast-DueAccounts'
-NODUPLICATE/NODUP/NODUPRECSoption
Inthisexample,theNODUPLICATEoptionremovesobservationsthathaveduplicatevalueswithBYvalue.
datasorting_6;
inputpat_idage;
29015666
28025664
28015664
26215658
26325655
procsortdata=sorting_6noduplicateout=sorting_7;
byage;
***Nodupkey,nodup,andnoduplicate***;
datadup;
inputaccount_idvisitmmddyy10.checking_balcomma9.2;
20118911/11/19987,865.28
20118911/28/19985,724.02
20118912/08/19986,908.98
20236911/11/19984,405.18
20418911/28/19985,724.02
20418912/05/19988,054.32
22518911/28/19983,632.85
225189
升级会员 