SAS程序范例1.docx

SAS程序范例1.docx

《SAS程序范例1.docx》由会员分享，可在线阅读，更多相关《SAS程序范例1.docx（20页珍藏版）》请在冰豆网上搜索。

SAS程序范例1

5.9历年农村家庭出售畜产品水产品因子分析

SAS数据集d1是历年农村家庭出售畜产品水产品（单位：

公斤）。

SAS数据集d1中，x1是猪肉，x2是牛肉，x3是羊肉，x4是家禽，x5是蛋类，x6是牛羊奶，x7是蚕茧，x8是水产品。

本例用Factor过程对SAS数据集进行因子分析。

Method是因子分析方法，本例用主分量法（Prin）；Rotate是因子旋转方法，本例用方差最大旋转方法（Varimax）。

%letd1=fjc.njcps;

%letc1=x1x2x3x4x5x6x7x8;

%leti1=date;

%macrofactor（method,rotate,x）;

%doi=1%to1;

title"factor&&d&imethod=&methodrotate=&rotatenfactors=&x";

procfactordata=&&d&imethod=&methodnfactors=&xscree

rotate=&rotateout=a&i;

var&&c&i;

procprintdata=a&i（keep=datefactor1factor2factor3factor4factor5）;

run;

%letplotitop=gopts=cback=blue,color=white,cframe=yellow;

%plotit（data=a&i,labelvar=&&i&i,

plotvars=factor2factor1,colors=magenta）;

%end;

%mend;

%factor（prin,varimax,5）;

表5.9.1d1关于变量x1–x8的因子分析

TheFACTORProcedure

InitialFactorMethod:

PrincipalComponents

PriorCommunalityEstimates:

ONE

EigenvaluesoftheCorrelationMatrix:

Total=8Average=1

EigenvalueDifferenceProportionCumulative

16.622257515.869673080.82780.8278

20.752584430.324255260.09410.9219

30.428329170.329070870.05350.9754

40.099258290.042884280.01240.9878

50.056374020.033280870.00700.9949

60.023093150.009261420.00290.9977

70.013831730.009560010.00170.9995

80.004271710.00051.0000

5factorswillberetainedbytheNFACTORcriterion.

TheFACTORProcedure

InitialFactorMethod:

PrincipalComponents

FactorPattern

Factor1Factor2Factor3Factor4Factor5

x10.905280.008660.412220.05426-0.03823

x20.97855-0.13329-0.053740.09498-0.01536

x30.92755-0.271120.193650.107810.10219

x40.981310.00391-0.15594-0.061620.04219

x50.968560.016920.00737-0.222690.09560

x60.88951-0.05390-0.438790.10359-0.01067

x70.585550.807890.025010.054770.01280

x80.97450-0.073130.02191-0.09248-0.18176

VarianceExplainedbyEachFactor

Factor1Factor2Factor3Factor4Factor5

6.62225750.75258440.42832920.09925830.0563740

FinalCommunalityEstimates:

Total=7.958803

x1x2x3x4x5x6x7x8

0.993929020.987473270.993428580.992876970.997183650.997513850.999336060.99706201

表5.9.2d1用最大方差旋转法的因子分析

TheFACTORProcedure

RotationMethod:

Varimax

OrthogonalTransformationMatrix

12345

10.672770.647350.325580.139950.05228

2-0.27899-0.188270.941270.013500.02345

30.67779-0.732930.05375-0.001600.02273

40.100450.088530.06986-0.91644-0.37057

50.00735-0.021620.015580.37465-0.92676

RotatedFactorPattern

Factor1Factor2Factor3Factor4Factor5

x10.891190.287910.328250.062090.07223

x20.668520.706690.196640.042430.02586

x30.842500.516900.066330.06532-0.08811

x40.547530.742440.311150.209910.03159

x50.630230.596640.317600.375660.04513

x60.326380.916980.222360.025530.00677

x70.191090.213190.956450.047410.01797

x80.680230.624300.240330.152010.25245

VarianceExplainedbyEachFactor

Factor1Factor2Factor3Factor4Factor5

3.25368473.03272001.37049880.22111320.0807867

FinalCommunalityEstimates:

Total=7.958803

X1X2X3X4X5X6X7X8

0.993929020.987473270.993428580.992876970.997183650.997513850.999336060.99706201

TheFACTORProcedure

RotationMethod:

Varimax

ScoringCoefficientsEstimatedbyRegression

SquaredMultipleCorrelationsoftheVariableswithEachFactor

Factor1Factor2Factor3Factor4Factor5

1.00000001.00000001.00000001.00000001.0000000

StandardizedScoringCoefficients

Factor1Factor2Factor3Factor4Factor5

x10.79098227-0.55597470.13469736-0.7373550.45526125

x20.157899150.31157218-0.0627429-0.9606059-0.1013109

x30.62360717-0.115901-0.1650845-0.3022728-2.0732228

x4-0.20537340.290640570.001858120.87071991-0.4639014

x5-0.1090969-0.1574243-0.06061252.71209425-0.7316066

x6-0.48057280.94775021-0.0087927-1.0078485-0.2292305

x7-0.1433306-0.14372711.08444468-0.3938737-0.3837445

x80.043478860.06328619-0.156105-0.33486873.33992889

表5.9.3d1因子分析结果

ObsdateFactor1Factor2Factor3Factor4Factor5

11985-1.10911-0.25369-2.181501.34490-0.07697

21989-1.32763-0.10876-0.956050.386680.13439

31990-1.23713-0.18648-0.34555-0.41269-0.22633

41991-0.93756-0.279030.14972-0.008630.13111

51992-0.85212-0.504130.575100.133130.54129

61993-0.34236-0.816860.94898-0.492730.11206

71994-0.48363-0.871271.65538-0.17423-0.75094

81995-0.32079-0.866821.333700.06509-0.79746

919960.99320-0.22694-1.26173-2.16888-1.47536

1019971.08633-0.19954-1.11979-0.78386-0.99129

111998-0.33411-0.00303-0.02264-0.557991.36122

1219990.97092-0.24442-0.45168-1.048652.54327

1320001.45850-0.64049-0.400712.44493-0.44471

1420011.43904-0.234510.812610.45093-0.65838

1520021.302920.465210.455250.575721.40237

1620030.203242.175270.283750.51298-0.41630

172004-0.509712.795490.52515-0.26668-0.38796

图5.9.1d1的（Factor2，Factor1）图（标号年份）

5.10地区城镇居民平均每人全年家庭收入因子分析

SAS数据集d1是各地区城镇居民平均每人全年家庭收入来源（2004年）。

x1是可支配收入，x2是总收入，x3是工薪收入，x4是经营净收入，x5是财产性收入，x6是转移性收入。

本例用Factor过程对SAS数据集进行因子分析。

Method是因子分析方法，本例用主分量法（Prin）；Rotate是因子旋转方法，本例用方差最大旋转方法（Varimax）。

“%plotit（data=b&i,labelvar=x1,

plotvars=factor1x1,colors=magenta）;

%plotit（data=b&i,labelvar=&&i&i,

plotvars=factor1x1,colors=magenta）;”

表示画出b&i的两个图，标号分别x1为与地区。

%letd1=fjc.dqrjsr;

%letc1=x1x2x3x4x5x6;

%leti1=d;

%macrofactor（method,rotate,x）;

%doi=1%to1;

title"factor&&d&imethod=&methodrotate=&rotatenfactors=&x";

procfactordata=&&d&imethod=&methodnfactors=&xscree

rotate=&rotateout=a&i;

var&&c&i;

procsortdata=a&iout=b&i;

byfactor1;

procprintdata=b&i（keep=dfactor1factor2factor3）;

run;

%letplotitop=gopts=cback=blue,color=white,cframe=yellow;

%plotit（data=b&i,labelvar=x1,

plotvars=factor1x1,colors=magenta）;

%plotit（data=b&i,labelvar=&&i&i,

plotvars=factor1x1,colors=magenta）;

%end;

%mend;

%factor（prin,varimax,3）;

表5.10.1d1关于变量x1–x6的因子分析

TheFACTORProcedure

InitialFactorMethod:

PrincipalComponents

PriorCommunalityEstimates:

ONE

EigenvaluesoftheCorrelationMatrix:

Total=6Average=1

EigenvalueDifferenceProportionCumulative

13.967828882.893271010.66130.6613

21.074557860.541006080.17910.8404

30.533551780.110441100.08890.9293

40.423110680.422159870.07050.9998

50.000950810.000950810.00021.0000

60.000000000.00001.0000

3factorswillberetainedbytheNFACTORcriterion.

FactorPattern

Factor1Factor2Factor3

x10.98304-0.16578-0.00417

x20.97755-0.19264-0.03349

x30.87138-0.34573-0.30864

x40.451230.809660.06503

x50.686380.48148-0.27725

x60.78219-0.055320.59670

VarianceExplainedbyEachFactor

Factor1Factor2Factor3

3.96782891.07455790.5335518

FinalCommunalityEstimates:

Total=5.575939

x1x2x3x4x5x6

0.993862880.993839960.974096680.863394560.779808780.97093565

表5.10.2d1用最大方差旋转法的因子分析

TheFACTORProcedure

RotationMethod:

Varimax

OrthogonalTransformationMatrix

123

10.789410.402020.46391

2-0.417340.90568-0.07467

3-0.45017-0.134660.88273

RotatedFactorPattern

Factor1Factor2Factor3

x10.847080.245620.46474

x20.867160.223030.43832

x30.971110.078760.15762

x4-0.010980.905940.20628

x50.465710.749330.03773

x60.371940.183990.89373

VarianceExplainedbyEachFactor

Factor1Factor2Factor3

2.76791571.53234861.2756742

FinalCommunalityEstimates:

Total=5.575939

x1x2x3x4x5x6

0.993862880.993839960.974096680.863394560.779808780.97093565

TheFACTORProcedure

RotationMethod:

Varimax

ScoringCoefficientsEstimatedbyRegression

SquaredMultipleCorrelationsoftheVariableswithEachFactor

Factor1Factor2Factor3

1.00000001.00000001.0000000

StandardizedScoringCoefficients

Factor1Factor2Factor3

x10.26347368-0.0390740.11958235

x2-0.6691815-0.40436983.27897357

x31.291425440.13631122-2.7841675

x4-0.19802020.74119647-0.1663793

x50.216483160.55725461-0.5213599

x6000

表5.10.3d1因子分析结果（按Factor1排序）

ObsdFactor1Factor2Factor3

1Heilongjiang-1.171590.061340.47570

2Jilin-1.079440.693070.19549

3Guizhou-1.015860.049990.00199

4Ningxia-0.98302-0.188360.23796

5Qinghai-0.89348-0.748930.55922

6InnerMongolia-0.759030.74000-0.19218

7Henan-0.72128-0.060940.12705

8Anhui-0.653670.17482-0.31600

9Jiangxi-0.64680-0.21335-0.18477

10Liaoning-0.57474-0.667950.54119

11Hebei-0.57013-0.469040.29540

12Sichuan-0.535940.34736-0.21636

13Gansu-0.36463-0.69588-0.50456

14Jiangsu-0.344690.943881.21347

15Shaanxi-0.29071-0.64101-0.33783

16Xinjiang-0.23192-0.53125-0.72922

17Tianjin-0.15457-0.063262.12146

18Hunan-0.13430-0.12676-0.33590

19Hainan-0.127420.23593-0.79185

20Yunnan-0.080611.19116-0.35119

21Shanxi-0.07836-0.24479-0.76952

22Hubei-0.06154-0.49023-0.60043

23Guangxi0.09846-0.15394-0.41539

24Chongqing0.21982-0.912880.13585

25Fujian0.518921.625520.05943

26Shandong0.82041-0.62905-1.10312

27Zhejiang1.215503.237090.33195

28Tibet1.79656-1.94695-2.48714

29Beijing1.95068-1.674162.70696

30Guangdong2.308231.57961-1.41059

31Shanghai2.54515-0.421021.74293

图5.10.1d1的（Factor2，X1）图（标号X1）

5.11地区生产总值因子分析

SAS数据集是地区生产总值（2004年），共有变量19个，分为4组。

第一组变量x1-x11；x1是地区生产总值（单位：

亿元），x2是第一产业，x3是第二产业，x4是工业，x5是建筑业，x6是第三产业，x7是交通运输仓储邮电通信业，x8是批发和零售贸易餐饮业，x9是金融，保险业，x10是房地产业，x11是其他行业。

第二组变量x12-x14，代表地区生产总值构成（%）；x12是第一产业，x13是第二产业，x14是第三产业。

第三组变量x15-x18，代表地区生产总值指数（上年=100）；x15是地区生产总值，x16是第一产业，x17是第二产业，x18是第三产业。

第四组是变量x19，为人均地区生产总值（元/人）。

SAS数据集d1对变量x1-x11用Factor过程对SAS数据集进行因子分析；SAS数据集d2对变量x1x2x3x6用Factor过程对SAS数据集进行因子分析。

Method是因子分析方法，本例用主分量法（Prin）；Rotate是因子旋转方法，本例用方差最大旋转方法（Varimax）。

“%plotit（data=b&i,labelvar=x1,

plotvars=factor1x1,colors=magenta）;

%plotit（data=b&i,labelvar=&&i&i,

plotvars=factor1x1,colors=magenta）;”

表示画出b&i的两个（Factor1，x1）图，标号分别x1为与地区。

“%plotit（data=b&i,la