一般可以用"objectname$p.value"的格式提取出p-value, 这时的p-value应该是精确的。确切起见，你可以把你具体的程序贴上来看看。

大虾。好像不行欸。。我打“print(aa$result$Perm.P)”。。。。出来的是NULL。
原句如下：

> library(mbmdr)
Loading required package: logistf
Loading required package: mice
Loading required package: Rcpp
Loading required package: lattice
mice 2.22 2014-06-10
Loading required package: mgcv
Loading required package: nlme
This is mgcv 1.8-4. For overview type 'help("mgcv-package")'.
mbmdr 2.6
>
>
> fan<-read.csv("ldlc-data.csv",header=TRUE)
> attach(fan)
> pp <- mbmdr(y=ldlcj_class414, data=fan[1:10], order=3, covar=fan[11:13], exclude=NA, risk.threshold=0.1,
+ output=NULL, adjust="covariates",family=binomial(link=logit),use.logistf=TRUE,
+ first.model=NULL, list.models=NULL,printStep1=FALSE)
>
>
> #Permutation test for all models with MIN.P <= 0.05
> order <- 3
> models <- subset(pp$result, select = 1:order)
> aa<-mbmdr.PermTest(pp,100,models)
> print(aa)