SAS PROC MCMC中的MISSING=CCMODELY

SAS/STAT® 14.3 User’s Guide The MCMC Procedure 中提到了 missing选项：

 If you specify MISSING=CC (CC stands for complete cases), PROC MCMC discards all observations
that have missing or partial missing values before carrying out the simulation.
 If you specify MISSING=AC (AC stands for all cases), PROC MCMC neither discards any missing
values nor augments them.
 If you specify MISSING=CCMODELY, PROC MCMC treats missing response values as parameters
and incorporates them as part of the simulation. But the procedure discards all observations that have
missing covariates. A covariate is a data set variable that appears in the program but does not appear to
the left of the tilde in the MODEL statement.
 If you specify MISSING=ACMODELY, PROC MCMC samples missing responses without discarding
observations that contain missing values in covariates.



有四种模式，MISSING=ACMODELY 可以比较好的处理缺失值，但是写了个程序却发现系统不识别MISSING=ACMODELY
程序如下：
proc mcmc data=dose outpost=doseout1 nmc=10000 thin=2 seed=0 MISSING=CCMODELY;
parms a b;
prior a  ~ uniform(-5,10);       
prior b  ~ uniform(-10,40);
p=logistic(a+b*dose);
model y ~ binomial(n,p);
run;


报错：
201  proc mcmc data=dose outpost=doseout1 nmc=10000 thin=2 seed=0 MISSING=CCMODELY;
                                                                          --------
                                                                          22
                                                                          76
ERROR 22-322: 语法错误，期望下列之一: ALLCASE, COMPLETECASE.
ERROR 76-322: 语法错误，语句将被忽略。

求大神帮助，谢谢！！

houcanni 发表于 2018-3-30 17:09
SAS/STAT® 14.3 User’s Guide The MCMC Procedure 中提到了 missing选项：

 If you specify MISS ...

最后发现是版本问题，SAS/STATA 13.2不支持这部分命令，14.3可以支持，尝试用university edition解决了