使用 rmcorr 包，我在使用 rmcorr_mat 时得到 NaN 值。当我使用 rmcorr 函数时，我得到 NaN 但还有一条警告消息
Warning message:
In sqrt(SSFactor/(SSFactor + SSresidual)) : NaNs produced
我可以用 0 代替 NaN 吗？ 对于某些人，我也得到了 0 的相关性。
有人可以解释一下 rmcorr 何时返回 NaN 和 0 的相关性吗？
谢谢
> =input_file$A001
[1] 0 0 5 0 0 0 9 3 10 13 0 0 43 38 8 6 38 11 22 7
> input_file$A002
[1] 0 0 0 0 11 25 0 0 6 11 0 0 0 0 0 0 0 0 5 6
>rmcorr(participant = factor(subject), measure1 = input_file$A001,measure2 = input_file$A002,CI.level = 0.95,CIs="analytic", dataset = input_file)
Repeated measures correlation
r
NaN
degrees of freedom
9
p-value
1
95% confidence interval
NaN NaN
Warning message:
In sqrt(SSFactor/(SSFactor + SSresidual)) : NaNs produced
> rmcorr(participant = factor(subject), measure1 = input_file$A003,measure2 = input_file$A004,CI.level = 0.95,CIs="analytic", dataset = input_file)
Repeated measures correlation
r
0
degrees of freedom
9
p-value
1
95% confidence interval
-0.599875 0.599875
> input_file$A003
[1] 11 3 2 16 0 0 10 2 0 0 0 0 8 18 0 0 0 0 0 0
> input_file$A004
[1] 0 0 0 0 0 0 0 0 33 106 0 0 0 0 0 45 48 99 0 151