用mice包进行缺失值插补时遇到问题

具体代码如下：

newdata<- read.csv('newdata.csv',header=T,sep=',')   #newdata 1740 obs. of 178 variables
str(newdata)

library(lattice)
library(MASS)
library(nnet)
library(mice)

F=c(4:21);for (i in F) newdata[, i] = as.numeric(newdata[, i])
F=c(22:178);for (i in F) newdata[, i] = factor(newdata[, i])
str(newdata)

is.na(newdata) 
n <- sum(is.na(newdata)) 
md.pattern(newdata)
imp<-mice(newdata,m=5,seed=1234)

报错

> imp<-mice(newdata,m=5,seed=1234)

 iter imp variable
  1   1  Richmond.RAS.ScaleError in solve.default(xtx + diag(pen)) : 
  system is computationally singular: reciprocal condition number = 3.317e-22

想求助一下各位大佬们，问题出在哪里？