Bootstrap code reporting no error or standard error
I wrote this code in R:
library(boot)
bs <- function(formula, data, indices) {
d <- data[indices,] # allows boot to select sample
fit <- lm(formula, data=d)
return(coef(fit))
}
results <- boot(data=z, statistic=bs,
R=1000, formula=z[,1]~z[,2])
I am trying to make a random x-sample using a dataframe for data that contains my response and my predictor, however my results are returned without bias and without std.
Bootstrap Statistics :
original bias std. error
t1* 83.5466254 0 0
t2* -0.6360426 0 0
Can anyone spot the problem?
source to share
Your formula is wrong. When you use z[,1]~z[,2]
, you are literally specifying a formula that has the first column z
as the answer and the second column z
as the independent variables. Note that it z
never changes. This is the parameter data=
that changes. Also, the formula syntax does not work with such positional indices. You need to use variable names. Here are some sample data
z <- data.frame(a=runif(50), b=runif(50))
Note how this doesn't work.
results <- boot(data=z, statistic=bs,
R=10, formula=z[,1]~z[,2])
results
# Bootstrap Statistics :
# original bias std. error
# t1* 0.45221233 0 0
# t2* 0.08818014 0 0
it just reconfigures the same values over and over, the same as when using the fill dataset
lm(a~b, z)
# Coefficients:
# (Intercept) b
# 0.45221 0.08818
Do you want to
results <- boot(data=z, statistic=bs,
R=10, formula=a~b)
results
# Bootstrap Statistics :
# original bias std. error
# t1* 0.45221233 0.01024794 0.08853861
# t2* 0.08818014 -0.01546608 0.16376128
This allows the function to boot
traverse on a different dataset each time, and since the vector values of the vector are not included in the formula that specifically references z
data.frame, you will get updated values.
source to share