样本数据集
df <- data.frame (species = rep(c("A","B","C","D"), times = 20),
factor = rep(c("1","2","3","4"), each = 10),
yaxis <- c(-31.71, 3.04, 2.24 , 86.67, 1.49 , 1.08, 1.90, 1.29, 1.24 , 2.21 , 3.01, 2.41 , 6.35 , -114.15 , 41.23 ,3.11 , 1.16 , 0.67 , -5.70 , 0.87 , 3.87 , -61.49 , 2.41 , 1.66 , 3.65 , 2.42 , 3.06 , 1.32, 2.03, 0.75 , 3.16, 1.90 , 4.77 , 0.10, 3.71 , -1.18 , 4.69 , 11.46 , 10.77 , 0.61 ,-26.54 , -0.21 , 47.89 , 1.42 -12.07 , 4.08 , 6.80 , 3.67 , 3.75 , 9.13 , 4.57 , 2.67 , 0.02 , 1.96 , -1.04 -0.61, 2.44 , -0.79 , 2.20 , 1.16 ,12.53 , 0.72 ,4.00 , 2.89 , 2.10 ,16.54 , 8.50 , 1.66 ,-15.02 ,-0.21 ,-5.29 , 3.12 ,-1.77 ,14.01 ,2.25 , 6.41 , 7.21 ,7.92 , 11.46 , 9.33, 0.03 , 3.43))
我想测试模型
Model : yaxis ~ species + (1|factor), data = df
底层分布:不明确(是一个有正负值的指数,无界)。我想引导数据并估计平均值和置信区间。我该怎么做?
我拼凑了几个答案并尝试了以下方法。我明白了
Error in statistic(data, original, ...) : could not find function "statistic"
library(boot)
my.function = y ~ x
sp <- split(df, df$species)
y <- lapply(sp, function(x){
avg <- mean(df$yaxis)
basic <- boot.ci(boot(x, my.function, R = 1000), type = "basic")$basic
CI.LL <- basic[4]
CI.UL <- basic[5]
data.frame(avg, CI.LL, CI.UL)
})
do.call(rbind, y)
嗨,你明白了吗?我遇到了同样的问题并陷入困境:<