R Packages

resample

Bootstrap, permutation tests, and other resampling functions, featuring easy-to-use syntax.

Announcement-only list; join this for notices of new versions: http://groups.google.com/group/r-resample-announce.

For a comparison of ease of use of boot and resample, see resamplePoster1407.pdf.

6/13/2022 - Version 0.6 is on CRAN, http://cran.fhcrc.org/web/packages/resample

Older: resample_0.3.tar.gz, for people reading What Teachers Should Know about the Bootstrap: Resampling in the Undergraduate Statistics Curriculum, Tim Hesterberg (2014), http://arxiv.org/abs/1411.5279

aggregate

Aggregation functions. Common aggregation tasks optimized for speed and ease of use.

11/23/2014 - I am not actively working on this package. R policies now prohibit .Call calls to routines internal to R, so I cannot put this on CRAN.

Announcement-only list; join this for notices of new versions: http://groups.google.com/group/r-aggregate-announce

Current version aggregate_1.3.0.tar.gz

dataframe

This contains versions of standard data frame functions in R, modified to avoid making extra copies of inputs. This is faster, particularly for large data.

11/23/2014 - I am not actively working on this package. (1) R has incorporated some of my improvements, so this package is less necessary. (2) I wouldn't be able to put new versions on CRAN, due to a change in CRAN policies (they now prohibit .Call calls to routines internal to R; these are present in some of the standard dataframe code that I modify).

Announcement-only list; join this for notices of new versions: http://groups.google.com/group/r-dataframe-announce (You can ignore the warning about this group containing adult content; it does not.)

Versions here:

dataframe_2.9.tar.gz. This is for R 3.0.2 (and probably later, but not tested). Remove all but SubsetRows and Math.data.frame.

The file memory.R (see below) has some benchmarks, including R 3.0.0 (without the dataframe package). R 3.0.0 does more large memory allocations than previous versions. They are not copies of the data. Some people from r-devel have looked into this, and made some improvements in R based on my findings.

Related talks

Seattle Area useR 5/23/2012: TwoPackages.pdf

useR! 2012, 6/15/2012: TwoPackages5.pdf

For a comparison of ease of use of boot and resample, see resamplePoster1407.pdf