0

目标：如何计算 r 中 Light 的 kappa 的引导置信区间？

问题：虽然我已成功提取统计信息，但我尝试了以下代码，但不确定我哪里出错了......

以下是数据集的示例：

head(event_judge)

 judge1  judge 2  judge3 
   0          1        1
   0          0        0
   1          1        1

这是我用来计算统计数据和 95% CI 的代码：


library(psy)
lkappa(event_judge) # Light's kappa for binary diagnosis

#to obtain a 95%confidence interval:

# ATTEMPT 1
library(boot)
lkappa.boot <- function(data,x) {lkappa(data[x,], type="weighted")}
res <- boot(event_judge[,1:3],lkappa.boot,1000)
#quantile(res$t,c(0.025,0.975)) # Bootstrapped confidence interval of Light's kappa
#boot.ci(res,type="bca") # adjusted bootstrap percentile (BCa) confidence interval (better)

在开始 res 的行之后，我收到以下错误： if ((sum(dim1[i] == dimi1) == 1) & (sum(dim2[j] == dimi2) == 中的错误：缺少值 where TRUE /FALSE 需要

# ATTEMPT 2
if (requireNamespace("psy")) {
    ## example comparable to that when called ?lkappa
    data("event_judge", package = "psy")
    set.seed(1)
    confIntKappa(dat = event_judge, type = "not Cohen", weights = "absolute", 
                 M = 200, conf.level = 0.95)
4

0 回答 0