WPT: A weighted permutation test for paired data with missing...

View source: R/WPT-function.R

WPTR Documentation

A weighted permutation test for paired data with missing values in both arms

Description

The WPT function calculates the weighted studentized permutation test for testing H0m: mu1 = mu2 for matched pairs with missingness in both arm. WPT test is a combination of the randomization version of the paired t-test (Konietschke and Pauly, 2014) and Janssen's permutation version for the Welch test (Janssen, 1997) for complate and incomplete data respectively.

Usage

WPT(
  x,
  y,
  alternative = c("two.sided", "less", "greater"),
  nperm = 10000,
  alpha = 0.05
)

Arguments

x

a (non-empty) numeric vector of data values representing the first components of the pairs.

y

a (non-empty) numeric vector of data values representing the second components of the pairs.

alternative

a character string specifying the alternative hypothesis, must be one of 'two.sided' (default), 'greater' or 'less'. You can specify just the initial letter.

nperm

The number of permutations used for calculating the weighted permutation test. The default option is 10,000.

alpha

A number specifying the significance level; the default is 0.05.

Value

A WPT object containing the following components:

Input

The data input of the user.

Descriptive

Some descriptive statistics of the data. Displayed are the number of individuals per variable, the mean, and variance.

WPT

The value of the weighted test staistic and the p-value of the permutation procedure, as well as the test decision of the weighted permutation test.

References

Janssen, A. (1997). Studentized permutation tests for non-iid hypotheses and the generalized Behrens-Fisher problem. Statistics & probability letters, 36(1), 9-21.

Konietschke, F., & Pauly, M. (2014). Bootstrapping and permuting paired t-test type statistics. Statistics and Computing, 24(3), 283-296.

Amro, L., and Pauly, M. (2017). Permuting incomplete paired data: a novel exact and asymptotic correct randomization test. Journal of Statistical Computation and Simulation, 87(6), 1148-1159.


lubnaamro/MissPair documentation built on Sept. 30, 2023, 3:47 a.m.