PBT: Parametric bootstrap tests for paired data with missing...

In lubnaamro/MissPair: Tests for Matched Pairs Designs with Missing Values

Parametric bootstrap tests for paired data with missing values in a single arm

Description

The PBT function calculates the asymptotic model based bootstrap version of three different quadratic forms for testing H0m: mu1 = mu2; the Wald-type statistic, the ANOVA-type statistic, and the MATS-type statistic for paired data with missingness in a single arm.

Usage

PBT(x, y, nbsp = 1000, 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.
nbsp	The number of bootstraps used for calculating the bootstrapped tests. The default option is 1000.
alpha	A number specifying the significance level; the default is 0.05.

Value

A PBT 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.
PBT	The value of the Wald-type statistic, the ANOVA-type statistic, and the MATS-type staistic and the p-value of the bootstrapped procedure of each test.

References

Amro, L., Pauly, M., and Ramosaj, B. (2019). Asymptotic based bootstrap approach for matched pairs with missingness in a single-arm.

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