(完整)SAS程序汇总

目录SAS基础..........................................................................................................................................2创建数据集...............................................................................................................................2其他软件产生的标准格式文件与SAS数据集之间的互相转换。..............................2从外部文件(文本文件)读取数据..................................................................................2排序...................................................................................................................................4计算产生新变量...............................................................................................................4数据集的拆分...........................................................................................................................4数据集的合并...........................................................................................................................5纵向连接...........................................................................................................................5横向合并...........................................................................................................................5计量资料的描述...............................................................................................................................6计算几何均数...........................................................................................................................6制作频数...............................................................................................................................6单变量描述...............................................................................................................................7多变量描述...............................................................................................................................8两样本均数的比较...........................................................................................................................9单一总体均数的可信区间.......................................................................................................9两总体均数相差的可信区间...................................................................................................9单样本均数的t检验..............................................................................................................10原始数据已知-ttest..................................................................................................10原始数据未知.................................................................................................................10配对资料两样本均数比较的t检验......................................................................................11MEANS.............................................................................................................................12UNIVARIATE.................................................................................................................12TTest.............................................................................................................................12成组资料两样本均数比较的t检验......................................................................................13两样本均数的等效性检验.....................................................................................................13完全随机设计资料的方差分析.....................................................................................................14相关................................................................................................................................................15绘制散点图.............................................................................................................................15绘制分层散点图.....................................................................................................................17Pearson相关系数及其置信区间的计算...............................................................................17Spearman相关系数的计算....................................................................................................18简单线性回归.................................................................................................................................18简单线性回归方程的估计及假设检验.................................................................................18绘制回归直线置信带、预测带.............................................................................................19根据输出的残差数据绘制残差图.........................................................................................19多变量回归分析.............................................................................................................................20卡方检验.........................................................................................................................................21独立四格表资料的卡方检验.................................................................................................21配对四格表资料率的比较的卡方检验.................................................................................22R*C表资料的卡方检验........................................................................................................231关联性分析.............................................................................................................................24基于秩次的非参数检验.................................................................................................................25配对样本的符号秩和检验.....................................................................................................25单样本的符号秩和检验.........................................................................................................25两独立样本的秩和检验.........................................................................................................26多组独立样本的秩和检验.....................................................................................................28随机区组样本的秩和检验.....................................................................................................28Logistic回归与生存分析...............................................................................................................29Logistic回归..........................................................................................................................29二分类Logistic回归......................................................................................................29有序LOGISTIC回归.....................................................................................................31多分类LOGISTIC回归.................................................................................................31生存分析.................................................................................................................................32寿命表进行生存率的估计.....................................................................................32K-M法进行生存率的估计............................................................................................33Cox回归.........................................................................................................................34SAS基础创建数据集其他软件产生的标准格式文件与SAS数据集之间的互相转换。libnamec"F:\郑亚_32620151151279";Procimportdatafile='E:\data1_1.xls'dbms=excelreplaceout=c.data1_1_32620151151279;/*sheet=”sheet1$”*/Run;从外部文件(文本文件)读取数据DATAb;INFILE'E:\SYS.TXT';INPUTIDGENDER$AGEBLOOD$SURT;RUN;DATAA;INPUTIDGENDER$AGEBLOOD$SURT;/*ID等为变量名，$表示其前面的那个变量为字符型变量*/2DATALINES;/*引出数据行*/1M41A3682M26B7453F35B4014M47AB5525F37A4786F39O6287M28O5498M31B1289M43AB46310M29A512;ODSHTMLFILE='E:\RESULT.xls';/*将以下SAS输出结果存放到EXCEL文件中*/PROCPRINTDATA=A;RUN;ODSHTMLCLOSE;/*关闭对外输出系统*/PROCFORMAT;VALUEGE1='M'2='F';/*定义值标签*/VALUEBL1='A'2='B'3='AB'4='O';DATAA;INPUTIDGENDERAGEBLOODSURT;DATALINES;114113682126274532352401414735525237147862394628712845498131212891433463101291512;RUN;ODSHTMLFILE='E:\RESULT.xls';PROCPRINTDATA=A;FORMATGENDERGE.;/*借用值标签*/FORMATBLOODBL.;RUN;ODSHTMLCLOSE;3排序Procsortdata=aout=b;byage;Run;Procsortdata=aout=c;bydescendingage;Run;计算产生新变量DATAB；SETA；打开数据集B，并从数据集A读入数据。DATAD;FORMATSURT_Y5.3;SETA;SURT_Y=SURT/365;IFAGE>40THENAGROUP=1;ELSEAGROUP=2;RUN;数据集的拆分DATAA;INPUTIDGENDER$AGEBLOOD$SURT;DATALINES;/*引出数据行*/1M41A3682M26B7453F35B4014M47AB5525F37A4786F39O6287M28O5498M31B1289M43AB46310M29A512;DATABC;SETA;IFGENDER='F'THENOUTPUTB;ELSEOUTPUTC;RUN;DATAB(DROP=GENDER)C(DROP=BLOOD);SETA;RUN;4数据集的合并纵向连接datast1;inputxy@@;/*@@表示连续读入数据*/cards;120240;procprint;datast2;inputxy@@;cards;3804160;procprint;datast;setst1st2;procprint;run;横向合并dataa;inputnumx@@;cards;101121021410315；procsortout=aa;bynum;datab;inputnumy@@;cards;101100102134103145；procsortout=bb;bynum;datac;mergeaabb;bynum;procprint;run;5计量资料的描述计算几何均数dataex2_5;inputxf@@;y=log10(x);cards;1042034010801016011320156401412802;procmeansnoprint;vary;freqf;outputout=bmean=logmean;run;datac;setb;g=10**logmean;procprintdata=c;varg;run;制作频数表dataex2_1;inputx@@;low=3.07;dis=0.2;z=x-mod(x-low,dis);cards;3.964.234.423.595.124.024.323.724.764.164.614.263.774.204.363.074.893.974.283.644.664.0464.554.254.633.914.413.525.034.014.304.194.754.144.574.264.563.793.894.214.953.984.293.674.694.124.564.264.664.283.834.205.244.024.333.764.814.173.963.274.614.263.964.233.764.014.293.673.394.124.273.614.984.243.834.203.714.034.344.693.624.184.264.365.284.214.424.363.664.024.314.833.593.973.964.495.114.204.364.543.723.974.284.763.214.044.564.254.924.234.473.605.234.024.324.684.763.694.614.263.894.214.363.425.014.014.293.684.714.134.574.264.035.464.163.644.163.76;procfreq;tablesz;run;单变量描述/*用MEANS过程计算例数、均数、标准差、标准误*/procmeansdata=ex2_1;varx;run;/*用MEANS过程计算例数、均数、标准差、标准误和95%可信区间*/procmeansdata=ex2_1nmeanstdstderrcvclm;varx;run;/*用MEANS过程对频数表计算例数、均数、标准差、最小值和最大值*/dataprg4_4;inputxf@@;cards;3.1723.3733.57973.77143.97224.17304.37214.57154.77104.9765.1745.372;procmeans;freqf;varx;run;procmeansmaxdec=2data=prg4_4;freqf;varx;run;多变量描述/*用UNIVARIATE过程进行描述性统计*/procunivariatedata=ex2_1;freqf;varx;run;/*计算2.5%和97.5%分位数*/procunivariatedata=ex2_1;freqf;varx;outputout=pctpctlpre=ppctlpts=2.597.5;run;procprintdata=pct;run;/*作正态性检验、茎叶图、箱式图和正态概率图*/procunivariatedata=ex2_1Normalplot;varx;run;8两样本均数的比较单一总体均数的可信区间dataprg5_1;n=10;mean=166.95;std=3.64;t=tinv(0.975,n-1);in=t*std/sqrt(n);lclm=mean-in;uclm=mean+in;procprint;varlclmuclm;run;两总体均数相差的可信区间dataprg5_2;n1=29;n2=32;m1=20.10;m2=16.89;s1=7.02;s2=8.46;ss1=s1**2*(n1-1);ss2=s2**2*(n2-1);sc2=(ss1+ss2)/(n1+n2-2);se=sqrt(sc2*(1/n1+1/n2));t=tinv(0.975,n1+n2-2);lclm=(m1-m2)-t*se;uclm=(m1-m2)+t*se;procprint;vartselclmuclm;run;9单样本均数的t检验原始数据已知-ttestdataprg5_1;inputx@@;datalines;;Run;Procttesth0=140;VarxRun;原始数据未知某医生测量了36名从事铅作业男性工人的血红蛋白含量，算得其均数为130.83g/L，标准差为25.74g/L。问从事铅作业工人的血红蛋白是否不同于正常成年男性平均值140g/L？dataprg5_3;n=36;s_m=130.83;std=25.74;p_m=140;df=n-1;t=abs(s_m-p_m)/(std/sqrt(n));p=(1-probt(t,df))*2;10procprint;vartp;run;配对资料两样本均数比较的t检验dataprg5_2;n1=29;n2=32;m1=20.10;m2=16.89;s1=7.02;s2=8.46;ss1=s1**2*(n1-1);ss2=s2**2*(n2-1);sc2=(ss1+ss2)/(n1+n2-2);se=sqrt(sc2*(1/n1+1/n2));t=tinv(0.975,n1+n2-2);lclm=(m1-m2)-t*se;uclm=(m1-m2)+t*se;procprint;vartselclmuclm;run;dataprg5_3;n=36;s_m=130.83;std=25.74;p_m=140;df=n-1;t=abs(s_m-p_m)/(std/sqrt(n));p=(1-probt(t,df))*2;procprint;vartp;run;dataprg5_4;inputx@@;datalines;1717913578118175122105111140138132142140168113145128124134116129155135134136113119132;11procttesth0=140;varx;run;MEANS/*用MEANS作配对资料两个样本均数比较的t检验*/dataprg5_5;inputx1x2@@;d=x1-x2;cards;0.8400.5800.5910.5090.6740.5000.6320.3160.6870.3370.9780.5170.7500.4540.7300.5121.2000.9970.8700.506;procmeansnmeanstdstderrtprt;vard;run;UNIVARIATE/*用UNIVARIATE过程作配对资料两样本均数比较的t检验*/procunivariatedata=prg5_5;vard;run;TTest/*用TTest过程作配对资料两样本均数比较的t检验*/procttestdata=prg5_5;vard;run;12成组资料两样本均数比较的t检验dataprg5_6;inputxc@@;datalines;-0.71-5.61212.810.713.51415.817.11-0.512.51-1.611.71310.414.514.612.5161-1.413.726.52525.220.820.220.623.426.62-1.12623.82221.62222.221.223.121.72-22;procttest;varx;classc;run;dataprg5_7;inputx@@;if_n_<21thenc=1;elsec=2;cards;-0.70-5.602.002.800.703.504.005.807.10-0.502.50-1.601.703.000.404.504.602.506.00-1.403.706.505.005.200.800.200.603.406.60-1.106.003.802.001.602.002.201.203.101.70-2.00;procttest;varx;classc;run;两样本均数的等效性检验dataprg5_8;n1=102;n2=100;m1=0.87;m2=0.48;s1=0.89;13s2=0.82;delta=0.52;ss1=s1**2*(n1-1);ss2=s2**2*(n2-1);sc2=(ss1+ss2)/(n1+n2-2);se=sqrt(sc2*(1/n1+1/n2));t=(m1-m2-delta)/se;p=probt(t,n1+n2-2)*2;procprint;vartp;run;完全随机设计资料的方差分析/*完全随机设计资料的方差分析*/datapro_251;doc=1to3;doi=1to12;inputx@@;output;end;end;datalines;332.96297.64312.57295.47284.25307.97292.12244.61261.46286.46322.49282.42253.21235.87269.30258.90254.39200.87227.79237.05216.85238.03238.19243.49232.55217.71216.15220.72219.46247.47280.75196.01208.24198.41240.35219.56;run;odshtmlfile='G:\新建文件夹\result.xls';procanova;classc;modelx=c;meansc;meansc/lsdbonhovtest;run;odshtmlclose;14/*随机区组设计资料的方差分析*/dataproc_2541;doa=1to10;dob=1to3;inputx@@;output;end;end;datalines;2.212.914.252.322.644.563.153.674.331.863.293.892.562.453.781.982.744.622.373.154.712.883.443.563.052.613.773.422.864.23;run;odshtmlfile='G:\新建文件夹\result2.xls';procanova;classab;modelx=ab;meansb/dunnett('3');/*meansb/snk*/run;odshtmlclose;相关绘制散点图libnamec"E:\20121225\教学\2014-2015第一学期\研究生\SAS应用\二、相关与回归分析";/*定义关联库*/datac.t1;labelx='Height(cm)'y='Weight(Kg)';inputxy@@;datalines;149.430.8167.642.6146.333.1170.744.015161.536.3164.640.8155.532.7158.535.4149.433.1152.431.8;/*建立数据集并输入数据*//*filenameoutput"E:\20121225\教学\2014-2015第一学期\研究生\SAS应用\二、相关与回归分析\t1scatter.jpeg";*/Goptionshtext=0.25inctext=blackdevice=gif;/*************************************//*定义绘图区格式*//*'htext=0.25in'定义字体大小0.25inche*//*'ctext=black'定义字体颜色*//*************************************/AXIS1label=(h=0.25inf=zapfbij=r"HEIGHT(cm)")offset=(0.2in)order=(145to175by10);AXIS2label=(a=90h=0.25inf=zapfbij=l"WEIGHT(Kg)")order=(30to45by5)offset=(0.2in);/*************************************//*定义坐标轴的格式*//*'h=0.25in'定义坐标轴标目的字体大小0.25inche*//*'f=zapfbi'定义坐标轴标目的字体zapfbi*//*'j=r'定义坐标轴标目靠右对齐*//*'HEIGHT(cm)'定义坐标轴标目为'HEIGHT(cm)'*//*'offset=0.2in'定义坐标轴起点处空0.2inche*//*'order=(145to175by10)'定义坐标轴刻度*//*************************************/Odshtmlpath="d:\"body="散点图.xls";Procgplotdata=c.t1;Ploty*x/HAXIS=AXIS1VAXIS=AXIS2noframe;Symbolh=0.25inc=black;run;/*************************************//*调用gplot过程绘制散点图*//*'plot'定义变量*//*'Haxis'定义横坐标轴的格式*//*'Vaxis'定义纵坐标轴的格式*//*'noframe'定义散点图无框*//*'SYMBOL'定义数据点的格式*//*'h=0.25in'定义数据点的大小为0.25inche*//*'c=black'定义数据点的颜色为黑色*//*************************************/odshtmlclose;16绘制分层散点图datac.t2;labelx='Height(cm)'y='Weight(Kg)';inputxy@@;IF_N_<=5THENG=1;ELSEG=2;datalines;149.430.8167.642.6146.333.1170.744.0161.536.3164.640.8155.532.7158.535.4149.433.1152.431.8;/*建立具有分层变量的数据*/goptionshtext=0.25inctext=blackdevice=gif;AXIS1label=(h=0.25inf=zapfbij=r"HEIGHT(cm)")offset=(0.2in)order=(145to175by10);AXIS2label=(a=90h=0.25inf=zapfbij=r"WEIGHT(Kg)")order=(30to45by5)offset=(0.2in);odshtmlpath="d:\"body="分层散点图.xls";procgplotdata=c.t2;ploty*x=G/HAXIS=AXIS1VAXIS=AXIS2noframe;symbol1h=0.25inc=blackv=plus;symbol2h=0.25inc=redv=plus;run;/*************************************//*绘制分层的散点图*//*'y*x=G'定义以变量'G'为分层变量*//*用两个symbol语句定义两层数据点的格式*//*************************************/ODSHTMLCLOSE;Pearson相关系数及其置信区间的计算odsHTMLFILE='E:\20121225\教学\2014-2015第一学期\研究生\SAS应用\二、相关与回归分析\RESULT.xls';proccorrdata=c.t1fisher;varxy;run;ODSHTMLCLOSE;/*************************************//*计算Pearson相关系数并将结果输出到EXCEL文件中*/17/*'fisher'指定要计算FisherZ变换计算的总体相关系数的置信区间*//*'var'定义进行相关分析的两个变量*//*************************************/Spearman相关系数的计算datac.t3;labelx='血红蛋白含量'y='贫血体征';inputxy@@;cards;5.045.836.127.318.839.1311.1112.3113.5113.81;odsHTMLFILE='E:\20121225\教学\2014-2015第一学期\研究生\SAS应用\二、相关与回归分析\RESULT2.xls';proccorrdata=c.t3spearman;varxy;run;ODSHTMLCLOSE;/*************************************//*计算Spearman相关系数并将结果输出到EXCEL文件中*//*'spearman'指定要计算Spearman相关系数*//*'var'定义进行相关分析的两个变量*//*************************/简单线性回归简单线性回归方程的估计及假设检验odsHTMLFILE='E:\20121225\教学\2014-2015第一学期\研究生\SAS应用\二、相关与回归分析\RESULT3.xls';procregdata=c.t1;modely=x/stb;plotstudent.*predicted.;outputout=c.rp=yhatr=restudent=rslcl=lyucl=uylclm=lybaruclm=uybar;run;ODSHTMLCLOSE;quit;18/*************************************//*估计简单线性回归方程并将结果输出到EXCEL文件中*//*'model'定义回归方程的自变量和因变量*//*'stb'定义计算标准化回归系数*//*'plot'绘制散点图*//*'student.*predicted.'定义绘制以学生化残差为纵坐标、预测值为横坐标的散点图*//*'output'定义将部分统计量输出到一个新的数据集中*//*'out=c.r'定义将输出数据集的名称和位置中*//*'p'定义输出预测值并命名为变量"yhat"*//*'r'定义输出残差并命名为变量"re"*//*'student'定义输出学生化残差并命名为变量"rs"*//*'lcl'定义输出个体预测值的95%区间下限的并命名为变量"ly"*//*'ucl'定义输出个体预测值的95%区间上限的并命名为变量"uy"*//*'lclm'定义输出总体均数的95%置信区间下限的并命名为变量"lybar"*//*'uclm'定义输出总体均数的95%置信区间上限的并命名为变量"uybar"*//*************************************/绘制回归直线置信带、预测带goptionshtext=0.25inctext=blackdevice=gif;symbol1h=0.25inc=blacki=rlclm95v=plus;symbol2h=0.25inc=redi=rlcli95;AXIS1label=(h=0.25inf=zapfbij=r"HEIGHT(cm)")offset=(0.2in)order=(145to175by10);AXIS2label=(a=90h=0.25inf=zapfbij=r"WEIGHT(Kg)")order=(30to45by5)offset=(0.2in);odshtmlpath="d:\"body="回归图.xls";procgplotdata=c.t1;ploty*xy*x/HAXIS=AXIS1VAXIS=AXIS2noframeoverlay;run;/************************************************//*绘制带有回归直线、置信带、预测带的散点图*//*'i=rlclm95'定义绘制回归直线和95%的置信带*//*'i=rlcli95'定义绘制回归直线和95%的预测带*//*'overlay'定义将多个图放置在同一个坐标系中*/ODSHTMLCLOSE;根据输出的残差数据绘制残差图goptionshtext=0.25inctext=blackdevice=gif;symbol1h=0.25inc=blacki=n;19AXIS1label=(h=0.25inf=zapfbij=r)offset=(0.2in);AXIS2label=(a=90h=0.25inf=zapfbij=r)offset=(0.2in);odshtmlpath="d:\"body="残差图.xls";procgplotdata=c.r;plotrs*y/haxis=axis1vaxis=axis2;run;/**********************************//*根据输出的残差数据绘制残差图*//*'i=n'定义不输出任何拟合线*//**********************************/多变量回归分析libnamec"F:\SAS\三、多变量回归分析";datac.t1;inputBMIDYLEPFPGADI@@;cards;24.2210.05.7513.629.3624.145.010.217.416.0124.223.09.326.214.3126.454.019.315.119.0319.0315.02.5011.126.0825.222.38.657.617.4623.393.05.669.719.6227.223.08.548.620.3619.494.02.837.342.8225.936.07.218.915.9224.386.06.867.322.7626.9912.08.757.015.3419.032.93.227.731.0025.717.013.0713.58.0521.119.04.906.017.2828.414.08.9013.512.3123.325.03.546.730.2526.394.023.268.25.5924.342.04.517.224.2828.7310.019.056.98.5923.828.08.479.118.9427.4616.019.446.58.8922.8620.09.928.116.0827.9910.017.336.114.1024.4912.06.017.029.5028.412.014.596.811.7423.376.04.316.325.6430.691.522.068.15.1820.817.03.467.132.2629.393.020.567.56.12;goptionshtext=0.25inftext="宋体"ctext=black;odshtmlpath="F:\SAS课件\三、多变量回归分析"body="result.xls";procregdata=c.t1;modelADI=BMIDYLEPFPG/stb;plotstudent.*predicted.;outputout=c.rp=yhatr=restudent=rslcl=lyucl=uylclm=lybaruclm=uybar;run;ODSHTMLCLOSE;quit;20AXIS1label=(h=0.25inf=zapfbij=r)offset=(0.2in);AXIS2label=(a=90h=0.25inf=zapfbij=r)offset=(0.2in);procgplotdata=c.r;plotrs*yhat/haxis=axis1vaxis=axis2;symbol1h=0.25inc=blacki=nv=plus;run;odsHTMLFILE='F:\SAS课件\三、多变量回归分析\RESULT1.xls';procregdata=c.t1;modelADI=BMIDYLEPFPG/stbselection=forward;run;ODSHTMLCLOSE;quit;odsHTMLFILE='F:\SAS课件\三、多变量回归分析\RESULT2.xls';procregdata=c.t1;mm1:modelADI=BMIDYLEPFPG/stbselection=forward;mm2:modelADI=BMIDYLEPFPG/stbselection=backward;mm3:modelADI=BMIDYLEPFPG/stbselection=stepwise;mm4:modelADI=BMIDYLEPFPG/stbselection=rsquare;mm5:modelADI=BMIDYLEPFPG/stbselection=adjrsq;mm6:modelADI=BMIDYLEPFPG/stbselection=cp;run;ODSHTMLCLOSE;quit;/*sle=0.10sls=0.15*/卡方检验独立四格表资料的卡方检验libnamec"F:\SAS课件\四、卡方检验\";PROCFORMAT;VALUEG1='万拉法新'2='氟西汀';/*定义值标签*/VALUELx1='有效'2='无效';run;datac.t1;labelR='组别'C='疗效';dor=1to2;doc=1to2;inputf@@;output;end;21end;datalines;19111515;/*建立数据集并输入数据*/ODSHTMLFILE="F:\SAS课件\四、卡方检\RESULT1.XLS";PROCfreqDATA=c.t1;/*调用FREQ过程进行卡方检验*/FORMATrG.;/*借用值标签*/FORMATcLx.;tabler*c/chisqnopercentnocolexpected;/*table/tables后面定义要对哪些变量做交叉表，‘chisq’表示要输出卡方检验的结果*//*nopercent要求不要输出百分比，nocol要求不要输出列百分比，expected要求要输出理论频数*/weightf;/*定义按什么变量进行加权*/RUN;ODSHTMLCLOSE;Procimportdatafile='F:\SAS课件\四、卡方检验\clinicaltrail.xls'dbms=excelreplaceout=c.t2;Run;PROCFORMAT;VALUEG1='试验药'2='安慰剂';VALUEge1='男'2='女';run;ODSHTMLFILE="F:\SAS课件\四、卡方检验\RESULT2.XLS";PROCfreqDATA=c.t2;FORMATgroupG.;FORMATgenderge.;tablegroup*gender/chisqnopercentnocolexpected;RUN;ODSHTMLCLOSE;配对四格表资料率的比较的卡方检验PROCFORMAT;VALUEt1='阳性'2='阴性';run;datac.t3;labelR='甲法'C='乙法';dor=1to2;22doc=1to2;inputf@@;output;end;end;datalines;80103111;ODShtmlpath="F:\SAS课件\四、卡方检验\"body="RESULT3.XLS";PROCfreqDATA=c.t3;FORM