0

In R, I can use the boot function (from here) to calculate some estimate of a statistic. How can I read the std.error or that estimate into an R variable from the output of the boot function?

Take this simple example using the Boston data set from the ISL book

get_mean <- function(data, index) {
  return(mean(data[index]))
}

bootstrap_res <- boot(Boston$medv, get_mean, 1000)
bootstrap_res

This prints


ORDINARY NONPARAMETRIC BOOTSTRAP


Call:
boot(data = Boston$medv, statistic = get_mean, R = 1000)


Bootstrap Statistics :
    original      bias    std. error
t1* 22.53281 0.007650791   0.4106622

How can I get the 0.4106622 value into a variable in R code ?

For example I can use

mean_original <- bootstrap_res$t0

To read 22.53281 into the mean_original variable.

How can I read the reported std.error ? Looking at the other fields in the boot's return, I do not see an obvious way

> summary(bootstrap_res)
          Length Class  Mode     
t0           1   -none- numeric  
t         1000   -none- numeric  
R            1   -none- numeric  
data       506   -none- numeric  
seed       626   -none- numeric  
statistic    1   -none- function 
sim          1   -none- character
call         4   -none- call     
stype        1   -none- character
strata     506   -none- numeric  
weights    506   -none- numeric
StupidWolf
  • 45,075
  • 17
  • 40
  • 72
Hakan Baba
  • 1,897
  • 4
  • 21
  • 37

1 Answers1

2

Simply calculate it:

bootstrap_res
#Bootstrap Statistics :
#    original       bias    std. error
#t1* 22.53281 -0.008953755    0.399066

sd(bootstrap_res$t)
#[1] 0.399066
Roland
  • 127,288
  • 10
  • 191
  • 288