R：一侧截断平均值(R: one side truncated mean)

R：一侧截断平均值(R: one side truncated mean)

我是R和这个论坛的新手，很抱歉这些愚蠢的问题。

我有一组反应时间数据（每个参与者约200个响应），但在计算平均值之前，我想将每个参与者的10％最慢响应定义为异常值并将其删除。

例如：如果我有20个值，我想切断它们中最慢的2个（= 10％）...

我觉得在R中应该有一个简单的方法，但我不知道如何开始。

提前致谢！

I'm new to R and this forum so sorry for stupid questions.

I have a set of reaction time data (~200 responses per participant) but before calculating the mean I want to define the 10% slowest responses of each participant as outliers and cut them off.

For example: If i have 20 values I want to cut off the 2 slowest of them (=10%)...

I feel like there should be an easy way to do so in R, but I have no idea how to start.

Thanks in advance!

最满意答案

您可以使用分位数来帮助解决这个问题。 例如：

data <- c(1, 2, 3, 4, 5, 6, 7, 8, 9 ,10) mean(data) > 5.5 #Restrict to data that is above 10th percentile mean(data[data > quantile(data, .10)]) > 6

You could use quantile to help with this. For example:

data <- c(1, 2, 3, 4, 5, 6, 7, 8, 9 ,10) mean(data) > 5.5 #Restrict to data that is above 10th percentile mean(data[data > quantile(data, .10)]) > 6