Lavaan无法计算标准差

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

原数据不便贴出，先给出代码

> sem1<-'
+ mb=~mbc+mbn
+ soil=~soc+tn
+ climate=~map+mat
+ '
> fit<-cfa(sem1,estimator="MLR",data=grad)
Error in solve.default(E) :
  system is computationally singular: reciprocal condition number = 3.27078e-22
In addition: Warning messages:
1: In lav_data_full(data = data, group = group, group.label = group.label,  :
  lavaan WARNING: some observed variances are (at least) a factor 1000 times larger than others; use varTable(fit) to investigate
2: In estimateVCOV(lavaanModel, samplestats = lavaanSampleStats, options = lavaanOptions,  :
  lavaan WARNING: could not compute standard errors!

另，google组里也有人问这事，但是跟我的情况不太一样

还有，结构方程计算前需要如何处理数据？

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏0 回帖

关键词：无法计算 标准差 fit SEM 标准差

Nicolle 发表于 2014-2-12 08:57
You got a Singularity Issue which means that a matrix is mathematically singular iff its determinant ...

I got the same question,your words made a work.but it have a new question,while i operated the code,i got a new warnning: lavaan WARNING: fit measures not available if test = "none",which means there were no statistic test,only estimate value were calculated.my code is:
model3<-"pra=~i1+i2+i3
                    mor=~i4+i5+i6+i7+i8
                    cog=~i9+i10+i11
                    lgtm=~i1+i2+i3+i4+i5+i6+i7+i8+i9+i10+i11"
model3.fit<-cfa(model3,data=lgtm,do.fit=F)