proc sql noprint;
create table table3 as
select
a.no,
b.*,
m*s as y,
(no-m)*s as y1
from table1 as a,
     table2 as b
where a.gender = b.sex
and   a.time   = b.agedays
and   a.no     = b.no
;
quit;

谢谢亲~

大神 我想问下，我想用proc nlmixed 程序进行二分类反应变量的两水平模型分析，在这个语句中，模型的参数是不是要用proc glimmix 语句进行参数的估计？于是我用以下程序：
proc glimmix method=rspl data=A;
class people;
model lowweightWAZ=month motheragedegree fatheragedegree mothereducation fatherBMIdegree fathereducation income inhabitant miscarriage parity gravidity gestation gender gwgdegree delivery preparation drink exercise vomit vaginal smoke noise computer phone depressiondegree nightfeed feeding breastnumber sleep sleeptime activity cold diarrhea complementaryfeedingtime motherBMIdegree/S dist=binary link=logit ddfm=bw;
random month motheragedegree fatheragedegree mothereducation fatherBMIdegree fathereducation income inhabitant miscarriage parity gravidity gestation gender gwgdegree delivery preparation drink exercise vomit vaginal smoke noise computer phone depressiondegree nightfeed feeding breastnumber sleep sleeptime activity cold diarrhea complementaryfeedingtime motherBMIdegree/subject=people;
nloptions tech=nrridg;
run;
出来的结果显示，有点参数为0，那我在proc nlmixed语句中那些参数为0，设置参数值为0吗？
还是我写的程序有问题？谢谢指点~

还有 我想问：
我按下面程序：
proc nlmixed data=A;
parms b0=0.3619 b1=9.2584 b2=0.1223 b3=-0.7684 b4=-0.8946 b5=-0.5084 b6=-0.1055 b7=0.5279 b8=0.5699 b9=0.9933 b10=1.1098 b11=0.211 b12=3.2812 b13=-0.3272 b14=0.466 b15=-1.0409 b16=-0.5441 b17=22.462 b18=-0.5337 b19=0.1345 b20=0.4873 b21=0.5315 b22=10.3754 b23=-1.1663 b24=1.3985 b25=-1.2495 b26=-0.2036 b27=-0.2045 b28=0.05201 b29=-0.03028 b30=-0.04162 b31=-0.02524 b32=0.5906 b33=-0.2547 b34=-0.2493 b35=-0.2544 sigma=1;
y=b0+u+b1*month+b2*motheragedegree+b3*fatheragedegree+b4*mothereducation+b5*fatherBMIdegree+b6*fathereducation+b7*income+b8*inhabitant+b9*miscarriage+b10*parity+b11*gravidity+b12*gestation+b13*gender+b14*gwgdegree+b15*delivery+b16*preparation+b17*drink+b17*exercise+b19*vomit+b20*vaginal+b21*smoke+b22*noise+b23*computer+b24*phone+b25*depressiondegree+b26*nightfeed+b27*feeding+b28*breastnumber+b29*sleep+b30*sleeptime+b31*activity+b32*cold+b33*diarrhea+b34*complementaryfeedingtime+b35*motherBMIdegree;
p=1/(1+exp(-y));
model lowweightWAZ~binary(p);
random u~normal(0,sigma) subject=people;
title 'only slope is random';
run;
结果出现ERROR: Quadrature accuracy of 0.000100 could not be achieved with 31 points.  The achieved accuracy was 1.000000.
这一问题该如何解决？

还有，我想问，我按下面程序：
proc nlmixed data=A;
parms b0=0.3619 b1=9.2584 b2=0.1223 b3=-0.7684 b4=-0.8946 b5=-0.5084 b6=-0.1055 b7=0.5279 b8=0.5699 b9=0.9933 b10=1.1098 b11=0.211 b12=3.2812 b13=-0.3272 b14=0.466 b15=-1.0409 b16=-0.5441 b17=22.462 b18=-0.5337 b19=0.1345 b20=0.4873 b21=0.5315 b22=10.3754 b23=-1.1663 b24=1.3985 b25=-1.2495 b26=-0.2036 b27=-0.2045 b28=0.05201 b29=-0.03028 b30=-0.04162 b31=-0.02524 b32=0.5906 b33=-0.2547 b34=-0.2493 b35=-0.2544 sigma=1;
y=b0+u+b1*month+b2*motheragedegree+b3*fatheragedegree+b4*mothereducation+b5*fatherBMIdegree+b6*fathereducation+b7*income+b8*inhabitant+b9*miscarriage+b10*parity+b11*gravidity+b12*gestation+b13*gender+b14*gwgdegree+b15*delivery+b16*preparation+b17*drink+b17*exercise+b19*vomit+b20*vaginal+b21*smoke+b22*noise+b23*computer+b24*phone+b25*depressiondegree+b26*nightfeed+b27*feeding+b28*breastnumber+b29*sleep+b30*sleeptime+b31*activity+b32*cold+b33*diarrhea+b34*complementaryfeedingtime+b35*motherBMIdegree;
p=1/(1+exp(-y));
model lowweightWAZ~binary(p);
random u~normal(0,sigma) subject=people;
title 'only slope is random';
run;
跑出来的结果是这样的
ERROR: Quadrature accuracy of 0.000100 could not be achieved with 31 points.  The achieved accuracy was 1.000000.
我该怎样解决？

我也出现了一模一样的情况，请问楼主解决了吗？

我想请教下，SAS里面PROC  NLMIXED中PARMS怎样估计？

你得提前知道原始显示公式

我按下面程序操作
proc nlmixed data=A;
parms b0=1 b1=1 b2=1 b3=1 b4=1 b5=1 b6=1  b7=1 b8=1 b9=1 b10=1 b11=1 b12=1 b13=1 b14=1 b15=1 b16=1 b17=1 b18=1 b19=11 b20=-1 b21=1 b22=1 b23=1 b24=1 b25=1 b26=1 b27=1 b28=1 b29=1 b30=1 b31=1 b32=-1  b33=1 b34=1 sigma=1;
y=b0+u+b1*x1+b2*x2+b3*x3+b4*x4+b5*x5+b6*x6+b7*x7+b8*x8+b9*x9+b10*x10+b11*x11+b12*x12+b13*x13+b14*x14+b15*x15+b16*x16+b17*x17+b18*x18+b19*x19+b20*x20+b21*x21+b22*x22+b23*x23+b24*x24+b25*x25+b26*x26+b27*x27+b28*x28+b29*x29+b30*x30+b31*x31+b32*x32+b33*x33+b34*x34;
p=1/(1+exp(-y));
model dependent~binary(p);
random u~normal(0,sigma) subject=id;
run;
结果出现
NOTE: The parameter complementaryfeedingtime is assigned the default starting value of 1.0, because it is not assigned an initial value with the PARMS statement.
ERROR: Quadrature accuracy of 0.000100 could not be achieved with 31 points.  The achieved accuracy was 1.000000.
这是由于参数设置问题吗？我查询其他类似proc nlmixed的参数设置都是给出值了，我不知道怎样设置参数？能够向你请教吗？谢谢指点~