如何解决错误:无法评估模型框架内的分组因子 X:如果可能,尝试将分组因子显式添加到数据框中
我试图在 lme4 的占用率中拟合一个二元线性混合模型,作为 acd 和 N_days 的乘积。由于我的抽样设计,有很多嵌套因素我想作为嵌套随机效应进行拟合。
这是数据框的几行:
important_data
和结构:
X sample FirstOrder SecondOrder Site family occupancy acd N_days date
1 57 A_1_02/03/2013 A_1 A_580 A Staphylinidae 0 28.50894 3 02/03/2013
2 118 A_1_02/11/2011 A_1 A_580 A Staphylinidae 1 28.50894 5 02/11/2011
3 119 A_1_02/11/2011 A_1 A_580 A Staphylinidae 1 28.50894 3 02/11/2011
4 120 A_1_02/11/2011 A_1 A_580 A Staphylinidae 1 28.50894 3 02/11/2011
5 121 A_1_02/11/2011 A_1 A_580 A Staphylinidae 1 28.50894 3 02/11/2011
6 122 A_1_02/11/2011 A_1 A_580 A Staphylinidae 1 28.50894 3 02/11/2011
field_name unique_ID
1 34 NA
2 41 14587
3 53 14587
4 54 14588
5 55 14589
6 56 14590
我使用显式嵌套来拟合我的模型:
'data.frame': 18608 obs. of 12 variables:
$ X : int 57 118 119 120 121 122 123 124 125 126 ...
$ sample : Factor w/ 1763 levels "A_1_02/03/2013",..: 1 2 2 2 2 2 2 2 2 2 ...
$ FirstOrder : Factor w/ 576 levels "A_1","A_10","A_11",..: 1 1 1 1 1 1 1 1 1 1 ...
$ SecondOrder: Factor w/ 193 levels "A_580","A_581",..: 1 1 1 1 1 1 1 1 1 1 ...
$ Site : Factor w/ 17 levels "A","B","C","D",..: 1 1 1 1 1 1 1 1 1 1 ...
$ family : Factor w/ 59 levels "Aderidae","Anobiidae",..: 56 56 56 56 56 56 56 56 56 56 ...
$ occupancy : int 0 1 1 1 1 1 1 1 1 1 ...
$ acd : num 28.5 28.5 28.5 28.5 28.5 ...
$ N_days : int 3 5 3 3 3 3 3 3 3 3 ...
$ date : Factor w/ 64 levels "01/03/2013","01/08/2013",..: 4 5 5 5 5 5 5 5 5 5 ...
$ field_name : int 34 41 53 54 55 56 57 58 59 60 ...
$ unique_ID : int NA 14587 14587 14588 14589 14590 14591 14592 14593 14594 ...
但是当我运行模型时出现错误:
staph_model <- glmer(formula = occupancy~acd+N_days+(1|Site)+(1|Site:SecondOrder)+(1|Site:SecondOrder:FirstOrder)+(1|Site:SecondOrder:FirstOrder:sample),data = staphylinidae,family = binomial(link = "logit"))
关于为什么会发生这种情况,我找不到任何明确的答案?
版权声明:本文内容由互联网用户自发贡献,该文观点与技术仅代表作者本人。本站仅提供信息存储空间服务,不拥有所有权,不承担相关法律责任。如发现本站有涉嫌侵权/违法违规的内容, 请发送邮件至 dio@foxmail.com 举报,一经查实,本站将立刻删除。