R Programming/Bootstrap

Wikipedia has related information at Bootstrapping (statistics)

R Programming

R Basics
Data Management
Graphics
- ggplot
Publication quality output
Descriptive Statistics
Mathematics
Statistical Core Methods
- Maximum Likelihood
- Method of Moments
- Bayesian Methods
- Bootstrap
- Multiple Imputation
- Nonparametric Methods
Regression Models
Time Series
Factor Analysis
Classification
- Ordination
- Clustering
Network Analysis
High Performance Computing
- Profiling R code
- Parallel computing with R
Appendix
- Sources
- Index

boot package includes functions from the book Bootstrap Methods and Their Applications by A. C. Davison and D. V. Hinkley (1997, CUP)
bootstrap package.

Quick how-to

Do a boostrap of some data for some function (here, mean):

 b <- boot(data, function(data, id) { mean(data[id]) }, 1000)

Use this to compute a 90%-confidence interval:

 boot.ci(b, .9, type="norm")

↑Jump back a section

References

Instructions for the boot package: http://www.statmethods.net/advstats/bootstrapping.html
Sample using the boot package: http://druedin.com/2012/11/10/bootstrapping-in-r/

Previous: Bayesian Methods

Next: Nonparametric Methods

↑Jump back a section

Last modified on 9 December 2012, at 20:30