Bootstrap Program in R

Question

I'm having the following error in my bootstrap program in R. Does anyone know how to fix this?

> local({pkg <- select.list(sort(.packages(all.available = TRUE)),graphics=TRUE)
+ if(nchar(pkg)) library(pkg, character.only=TRUE)})

> library(boot)

> RH = c(20, 15, 15, 20, 20, 25, 30, 30, 35, 35, 40, 40, 40, 40, 45, 50, 50, 50, 60, 60, 70, 20, 30, 35, 35, 35, 35, 45, 45, 45, 45, 50, 50, 50, 50, 55, 65, 65, 65, 65, 75, 75, 80)
> 
> results <- boot(data=RH, mean, R=1000)


Error in mean.default(data, original, ...) : 

  'trim' must be numeric of length one

Thanks!

Answer 1

From ?boot your statistic requires two parameters, one for the data and one for the index.

This function calculates the sample mean based on the indices of the vector.

samp_mean <- function(x, i) {
  mean(x[i])
}

And the function call with samp_mean instead of mean

results <- boot(data=RH, samp_mean, R=1000)
results

ORDINARY NONPARAMETRIC BOOTSTRAP


Call:
boot(data = RH, statistic = samp_mean, R = 1000)


Bootstrap Statistics :
    original      bias    std. error
t1* 44.30233 -0.04430233    2.610505

Answer 2

The second argument to mean.default is trim, and boot is passing to mean a vector of indices as argument 'trim' which it is not expecting and quite rightly throws this error. Write a wrapper to mean, that accepts two arguments, the first as a data vector and the second as a permuted index, then use these to form a call to mean, here our wrapper function mean.fun does this (note we turn on removing NA's by default otherwise it wouldn't work)

library(boot)
RH <- c(20, 15, 15, 20, 20, 25, 30, 30, 35, 35, 40, 40, 40, 40, 45, 50, 50, 50, 60, 60, 70, 20, 30, 35, 35, 35, 35, 45, 45, 45, 45, 50, 50, 50, 50, 55, 65, 65, 65, 65, 75, 75, 80)
mean.fun <- function(dat, idx) mean(dat[idx], na.rm = TRUE) 
boot.result <- boot(RH, mean.fun, R=1000, sim="ordinary") 
boot.result

The Result:

ORDINARY NONPARAMETRIC BOOTSTRAP


Call:
boot(data = RH, statistic = mean.fun, R = 1000, sim = "ordinary")


Bootstrap Statistics :
    original    bias    std. error
t1* 44.30233 0.0694186    2.521452