Skip to content

charleskelley/comps

Repository files navigation

Comps

Comparison & Causal Analysis

The Comps package makes comparative analysis easy and supports causal-comparitive experimental design or observational analysis used for causal inference. Along these lines, the package has two primary objectives.

  1. Support general comparison across groups quickly in an opinionated manner where ease of use and domain specific formatting are preferred over customizability
  2. Facilitate causal analysis by providing algorithms for matching observations to create comparison groups and statistically quantifying incremental differences between those groups

The Comps package helps one easily balance attributes across sample groups drawn from a specific population or find the closest individual matches from one sample set within another population or sample set.

These two approaches, 1) matching or balancing groups and 2) matching individual observations are frequently used in isolation or in tandem for comparative analysis, observational studies, and experimental design. Therefore, it's extremely helpful to have a package that allows you to conveniently move between the approaches or combine them in fluid manner.

See the Comps documentation for complete details.

About

For making comparisons between stuff

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages