how can I can estimate the standard error of the 0.1 percentile using the bootstrap?

I have calculated an estimate for the tenth percentile of the column medv in the Boston data set. I want to calculate the standard error of the 0.1 percentile using the bootstrap. I have written this function to calculate it but It is not giving me the right value.

Can you show me a way to calculate the standard error of a certain percentile using bootstrapping?

Note: I'm aware of the package 'boot' but I want to use a custom function.

library(MASS)  data(Boston)  μ.hat0.1 <- quantile(medv, c(0.1))      set.seed(20218)  boot <- function(times, data=Boston$medv) {    boots <- rep(NA, times)    for (i in 1:times) {      boots[i] <- quantileSE(sample(data, length(data), replace=TRUE),p=0.1)    }    boots  }  mean(boot(times=1000))  

https://stackoverflow.com/questions/66775065/how-can-i-can-estimate-the-standard-error-of-the-0-1-percentile-using-the-bootst March 24, 2021 at 01:06PM