R help - Jan 2010 - Assistance with boot() Package

Good Evening R Community:

I believe I understand the basics of using the boot() bootstrap resampling
function in the boot() package. I have not had any trouble creating a
boot.object to which I apply the boot.ci() function to calculate one or all of
the available confidence intervals.

What I am not sure about is if this set of functions can generate more than one
confidence interval of one or all of the types available.

I have a large data set (n=133,456) data set from which I would like to remove
random samples of different sizes and then calculate 95% confidence intervals
for the mean, 10% trimmed mean and median. I would like to determine how often
the confidence intervals generated by boot.ci() contain the mean, 10% trimmed
mean and median of the large data set.

I have looked at some examples for using the boot() and boot.ci() functions to
generate confidence intervals for the intercept and predictive variables from a
regression model, but I do not, or cannot I suppose, determine how I can
generate more than one set of normal, basic, percentile and BCa confidence
intervals using these two functions.

I am running R version 2.9.2 on an IBM T61 laptop. My OS is Win XP professional
SP 3, and the machine has a 1.99 GHz processor with 2.99 GB of RAM. The version
of the boot() package I am running is 1.2-41.

Thanks in advance for taking the time to help me.

Adam


      
	[[alternative HTML version deleted]]

On Sun, 3 Jan 2010, Adam Carr wrote:
> Good Evening R Community:
>
> I believe I understand the basics of using the boot() bootstrap 
> resampling function in the boot() package. I have not had any 
> trouble creating a boot.object to which I apply the boot.ci() 
> function to calculate one or all of the available confidence 
> intervals.
>
> What I am not sure about is if this set of functions can generate 
> more than one confidence interval of one or all of the types 
> available.
>
> I have a large data set (n=133,456) data set from which I would like 
> to remove random samples of different sizes and then calculate 95% 
> confidence intervals for the mean, 10% trimmed mean and median. I 
> would like to determine how often the confidence intervals generated 
> by boot.ci() contain the mean, 10% trimmed mean and median of the 
> large data set.
>
> I have looked at some examples for using the boot() and boot.ci() 
> functions to generate confidence intervals for the intercept and 
> predictive variables from a regression model, but I do not, or 
> cannot I suppose, determine how I can generate more than one set of 
> normal, basic, percentile and BCa confidence intervals using these 
> two functions.
Well, it can be done easily.  Studying the book for which 'boot' is 
support software would be a good start, but a hint is to look at the 
'index' argument to boot.ci: basically boot() can be called with a 
'statistic' which returns a vector, then boot.ci() called on each of 
the components of interest.  There is an example in MASS (the book) on 
pp 225-6.
> I am running R version 2.9.2 on an IBM T61 laptop. My OS is Win XP 
> professional SP 3, and the machine has a 1.99 GHz processor with 
> 2.99 GB of RAM. The version of the boot() package I am running is 
> 1.2-41.
>
> Thanks in advance for taking the time to help me.
>
> Adam
>
>
>
> 	[[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
PLEASE do -- no HTML mail for a start, and please update your R.


-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595