| Type: | Package |
| Title: | Likelihood-Based Evidence Ratios for Classical Statistical Tests |
| Version: | 0.1.0 |
| License: | MIT + file LICENSE |
| Language: | en-GB |
| Description: | Implements likelihood-based evidence ratios for unified reporting in classical statistical testing. The package reports effect estimates, uncertainty intervals, and likelihood ratios on the log 10 scale derived from a single statistical model. It applies to standard normal mean tests, contingency tables, and regression coefficients, and provides a direct evidential measure while retaining classical error guarantees. For the Evidence Ratio Reporting Standard see Lawless (2026) <doi:10.5281/zenodo.18261076>. |
| Encoding: | UTF-8 |
| Depends: | R (≥ 4.0) |
| Imports: | stats |
| Suggests: | testthat (≥ 3.1.0), waldo (≥ 0.5.0), knitr, rmarkdown, ggplot2 |
| VignetteBuilder: | knitr |
| RoxygenNote: | 7.3.2 |
| NeedsCompilation: | no |
| Packaged: | 2026-01-16 12:33:45 UTC; akira |
| Author: | Dylan Lawless 
[aut, cre, cph] |
| Maintainer: | Dylan Lawless <admin@switzerlandomics.ch> |
| Repository: | CRAN |
| Date/Publication: | 2026-01-21 20:00:08 UTC |
Likelihood-based evidence ratio test
Description
Computes an effect estimate, a Wald-style uncertainty interval, and a likelihood-based evidence ratio (log10 scale) from a single statistical model.
Usage
evidence_test(...)
Arguments
... |
Arguments defining the data and model. |
Value
An object of class evidence_result.
Examples
x <- sleep$extra[sleep$group == 1]
evidence_test(x)
tbl <- matrix(c(30, 70, 20, 80), nrow = 2)
evidence_test(tbl)
evidence_test(mpg ~ wt, data = mtcars, coef = "wt")