| Type: | Package |
| Title: | Uniform Exact Functional Tests for Contingency Tables |
| Version: | 1.0.1 |
| Date: | 2025-02-20 |
| Author: | Yiyi Li  [aut,
cre],
Joe Song [aut] |
| Maintainer: | Yiyi Li <gtarex@nmsu.edu> |
| Description: | Testing whether two discrete variables have a functional relationship under null distributions where the two variables are statistically independent with fixed marginal counts. The fast enumeration algorithm was based on (Nguyen et al. 2020) <doi:10.24963/ijcai.2020/372>. |
| License: | LGPL (≥ 3) |
| Encoding: | UTF-8 |
| Imports: | Rcpp (≥ 1.0.5) |
| LinkingTo: | Rcpp |
| Depends: | R (≥ 3.5.0), stats |
| Suggests: | knitr, rmarkdown, testthat (≥ 3.0.0) |
| VignetteBuilder: | knitr |
| Config/testthat/edition: | 3 |
| RoxygenNote: | 7.3.2 |
| NeedsCompilation: | yes |
| Packaged: | 2025-02-21 06:36:22 UTC; gtarex |
| Repository: | CRAN |
| Date/Publication: | 2025-02-21 07:00:02 UTC |
Uniform Exact Functional Test on Two Discrete Random Variables
Description
Perform the uniform exact functional test on a contingency table to determine if the column variable is a function of the row variable.
Usage
UEFT(input, correct, log.p)
Arguments
input |
A matrix of nonnegative integers representing a contingency table. Column is the casual and row is the effect. |
correct |
Logical; if implement the continuity correction. The description is at details. The default is TRUE. |
log.p |
Logical; if TRUE, the p-value is given as log(p). The default is FALSE. The default is FALSE. |
Details
The uniform idea was implementated using uniform marginal distribution of a square table as null hypothesis.
Value
The exact p-value of the test.
Note
The functions provide a direct entry into the C++ implementations of the exact functional test.
Author(s)
Yiyi Li, Joe Song
Examples
# Initial a table
x = matrix(c(0,5,10,0,0,5), ncol=3)
# With continuity correction
UEFT(x)
# Without continuity correction
UEFT(x, FALSE)