統(tǒng)計(jì)軟件第13章外部文件

第13章外部文件程序自帶的數(shù)據(jù)從外部文本文件讀取數(shù)據(jù)INFILE選項(xiàng)從多個(gè)文件讀取數(shù)據(jù)把ASCII或原始數(shù)據(jù)寫(xiě)入外部文件寫(xiě)入CSV文件創(chuàng)建永久SAS數(shù)據(jù)集讀取永久SAS數(shù)據(jù)集如何查看SAS數(shù)據(jù)集的內(nèi)容帶有輸出格式的永久SAS數(shù)據(jù)集處理大型數(shù)據(jù)集程序自帶的數(shù)據(jù)DATAEX1;

INPUTGROUP$XYZ;DATALINES;CONTROL121719TREAT232529CONTROL191816TREAT222229;DATATEST;

INPUTAUTHOR$10.TITLE$40.;DATALINES4;SMITHTheUseofthe;inWritingFIELDCommentaryonSmith'sBook;;;;從外部文本文件讀取數(shù)據(jù)用INFILE語(yǔ)句省略DATALINES語(yǔ)句DATAEX2A;

INFILE

'F:\PPT\統(tǒng)計(jì)軟件\MYDATA.txt';INPUTGROUP$XYZ;RUN;PROC

MEANS

DATA=EX2AN

MEAN

STD

STDERR

MAXDEC=2;

VARXYZ;RUN;DATAEX2B;

FILENAMETRUMP'F:\PPT\統(tǒng)計(jì)軟件\MYDATA.txt';

INFILETRUMP;INPUTGROUP$XYZ;RUN;數(shù)據(jù)第一行有變量名，則在INFILE語(yǔ)句后添加選項(xiàng)FIRSTOBS=2INFILE選項(xiàng)*讀取幾個(gè)不同文件然后合并;DATAEX2E;

IFTESTENDNE1

THEN

INFILE

'F:\PPT\統(tǒng)計(jì)軟件\Cody_Program_Files\OSCAR.txt'

END=TESTEND;

ELSE

INFILE

'F:\PPT\統(tǒng)計(jì)軟件\BIGBIRD.TXT';

INPUTGROUP$XYZ;RUN;END=variablename選項(xiàng)TITLE;DATA

_NULL_;INFILE

'F:\PPT\統(tǒng)計(jì)軟件\Cody_Program_Files\CLINIC.DAT'

PAD

END=LAST;

INPUT@1ID$3.@4GENDER$1.@5DOBMMDDYY10.@15HR3.@18SBP3.@21DBP3.;

IFMISSING(GENDER)THENN_GENDER+1;

IFMISSING(DOB)THENN_DOB+1;

IFMISSING(HR)THENN_HR+1;

IFMISSING(SBP)THENN_SBP+1;

IFMISSING(DBP)THENN_DBP+1;

FILE

PRINT;

IFLASTTHEN

PUT

"SummaryReportofMissingValues"/

40*'-'

/

"NumberofMissingvaluesforGENDER:"N_GENDER/

"NumberofMissingValuesforDOB:"N_DOB/

"NumberofMissingValuesforHR:"N_HR/

"NumberofMissingValuesforSBP:"N_SBP/

"NumberofMissingValuesforDBP:"N_DBP;RUN;MISSOVER選項(xiàng)*記錄長(zhǎng)度不等或記錄末尾有缺失值;DATAEX2F;

INFILE

'F:\PPT\統(tǒng)計(jì)軟件\MYDATA1.txt'

MISSOVER;

INPUTGROUP$XYZ;RUN;CONTROL123TREAT45CONTROL678TREAT8910PAD選項(xiàng)*PAD選項(xiàng)可用于處理短記錄問(wèn)題;DATAEX2G;

INFILE

'F:\PPT\統(tǒng)計(jì)軟件\MYDATA2.TXT'

PAD;

INPUTGROUP$1X2-3Y4-5Z6-7;RUN;A123456B56A556677LRECL=record-length選項(xiàng)*INFILEfilerefLRECL=310;DATAEX2H;

INFILEDATALINESMISSOVER;

INPUTXYZ;DATALINES;12345678;從多個(gè)文件讀取數(shù)據(jù)DATAALL_THREE;

INFILE

'F:\PPT\統(tǒng)計(jì)軟件\Cody_Program_Files\FILE*.DAT'

MISSOVER

DLM=',';

INPUTXYZ;RUN;DATAREAD_MANY;

FILENAMEMIKE('F:\PPT\統(tǒng)計(jì)軟件\Cody_Program_Files\AAA.DAT'

'F:\PPT\統(tǒng)計(jì)軟件\Cody_Program_Files\BBB.DAT'

'F:\PPT\統(tǒng)計(jì)軟件\Cody_Program_Files\CCC.DAT');

INFILEMIKEMISSOVER;

INPUTXYZ;RUN;把ASCII或原始數(shù)據(jù)寫(xiě)入外部文件讀取——INFILE和INPUT語(yǔ)句寫(xiě)入——FILE和PUT語(yǔ)句DATAEX3A;

INFILE

'F:\PPT\統(tǒng)計(jì)軟件\MYDATA.txt';***Inputfile;

FILE

'F:\PPT\統(tǒng)計(jì)軟件\NEWDATA';***Outputfile;

INPUTGROUP$XYZ;TOTAL=SUM(OFXYZ);

PUTGROUP$1-10@12(XYZTOTAL)(5.);RUN;寫(xiě)入CSV文件*利用SAS的輸出設(shè)備系統(tǒng)（ODS）創(chuàng)建CSV文件;OPTIONS

MISSING="";DATACOMMA_DELIMITED;

INPUTNAME$XYZ;DATALINES;CODY123SMITH456MISS.89;ODS

LISTING

CLOSE;***Turnoflisting;ODS

CSV

FILE='F:\PPT\統(tǒng)計(jì)軟件\COMMA_ODS.CSV';PROC

PRINT

DATA=COMMA_DELIMITEDNOOBS;

TITLE;RUN;ODS

CSV

CLOSE;ODS

LISTING;創(chuàng)建永久SAS數(shù)據(jù)集LIBNAMEFELIX'F:\PPT\統(tǒng)計(jì)軟件\SASDATA';DATAFELIX.EX4A;

INPUTGROUP$XYZ;DATALINES;CONTROL121719TREAT232529CONTROL191816TREAT222229;讀取永久SAS數(shù)據(jù)集*LIBNAMEFELIX'F:\PPT\統(tǒng)計(jì)軟件\SASDATA';PROC

MEANS

DATA=FELIX.EX4AN

MEAN

STD

STDERR

MAXDEC=3;VARXYZ;RUN:如何查看SAS數(shù)據(jù)集的內(nèi)容PROC

CONTENTS

DATA=FELIX.EX4AVARNUM;

TITLE

"DemonstratingPROCCONTENTS";RUN:帶有輸出格式的永久SAS數(shù)據(jù)集LIBNAMEFELIX'F:\PPT\統(tǒng)計(jì)軟件\SASDATA';OPTIONS

FMTSEARCH=(FELIX);PROC

FORMAT

LIBRARY=FELIX;

VALUE$XGROUP'TREAT'='TREATMENTGRP'

'CONTROL'='CONTROLGRP';RUN;DATAFELIX.EX4A;

INPUTGROUP$XYZ;

FORMATGROUP$XGROUP.;DATALINES;CONTROL121719TREAT232529CONTROL191816TREAT222229;LIBNAMEC'F:\PPT\統(tǒng)計(jì)軟件\SASDATA';OPTIONS

FMTSEARCH=(C);PROC

PRINT

DATA=C.EX4A;RUN;處理大型數(shù)據(jù)集不要讀取不必要的文件丟棄不必要的變量SET語(yǔ)句中使用DROP或KEEP選項(xiàng)避免過(guò)多地對(duì)數(shù)據(jù)排序PROCMEANS中盡量用CLASS語(yǔ)句而不是BY語(yǔ)句選擇子集時(shí)用WHERE語(yǔ)句不用IF語(yǔ)句若只需要在數(shù)據(jù)的子集運(yùn)行某個(gè)過(guò)