The purpose of theprocs package is to recreatecommonly used SAS® procedures in R. This version of the package willattempt to simulate PROC FREQ, PROC MEANS, PROC TTEST, and PROC REG.Subsequent versions of the package will add more statisticalprocedures.
The package will also include a conceptual reproduction of PROCTRANSPOSE, PROC SORT, and PROC PRINT. The reason these procedures areincluded is because they are frequently used in conjunction with thestatistical procedures to prepare results for reporting.
There are three motivations for simulating SAS® statisticalprocedures in R:
1) Rich Outputs: R provides a wide range ofstatistical packages and functions. However, these statistical functionsfrequently return a fragmented output. The user is often left to compileR statistical results into a readable report themselves.
SAS® software, on the other hand, provides the same statisticalresults, but with rich outputs. These outputs include multiple datasets,plots, and complete statistical reports. These rich outputs could be abenefit to R users.
2) Pre-Validation: There is an industry-widemovement to adopt R for statistical analysis. This effort is complicatedby the fact that R statistical procedures frequently do not produceidentical output to SAS®. Many hours are burned trying to figure out whyR statistical results do not match SAS® statistical results.
Therefore another goal of the package is to match SAS® statisticswith as much fidelity as possible. If R statistical results match SAS®output, it makes it much easier to rewrite SAS® programs in R. Thispre-validation will ultimately save the industry tens of thousands ofhours of effort. Validation documentation is availablehere.
3) Ease of Adoption: Another important aspect of thepackage is to facilitate the adoption of R by SAS® programmers. SAS®programmers will easily understand and be comfortable with the functionsin this package. They will become productive in R much faster than withBase R functions, or statistical functions from other packages.
4) Stability: SAS® software is renowned for itsbackwards compatibility. A final goal of theprocspackage is to create a set of functions with a similarly intense concernfor stability. The package aims to avoid breaking changes at all costs.The dream is that programs written with these functions will still workyears from now.
The easiest way to install theprocs package is torun the following command from your R console:
install.packages("procs")Then put the following line at the top of your script:
library(procs)For examples and usage information, please visit theprocs documentation sitehere
Theprocs package is under active development. Ifyou want the latest development version, you can download from thegithub site by running the following code:
devtools::install_github("https://github.com/dbosak01/procs")If you need help, the first place to turn to is theprocs web site.
If you want to look at the code for theprocspackage, visit the github pagehere.
If you encounter a bug or have a feature request, please submit anissuehere.
Theprocs packages has been validated, and thevalidation documentation ishere.
Theprocs package is part of thesassy meta-package. Thesassymeta-package includes several packages that help make R easier. You canread more about thesassy packagehere.