trend.test: Test for increasing trend with correlated binary data

Description

The trend.test function provides a common interface to the trend tests implemented in this package: SO.trend.test, RS.trend.test, and GEE.trend.test. The details of each test can be found on their help page.

Usage

trend.test(
  cbdata,
  test = c("RS", "GEE", "GEEtrend", "GEEall", "SO"),
  exact = test == "SO",
  R = 100,
  control = soControl()
)

Value

A list with two components and an optional "boot" attribute that contains the detailed results of the permutation test as an object of class boot if an exact test was performed.

statistic: numeric, the value of the test statistic
p.val: numeric, asymptotic one-sided p-value of the test

Arguments

cbdata: a CBData object
test: character string defining the desired test statistic. "RS" performs the Rao-Scott test (RS.trend.test), "SO" performs the stochastic ordering test (SO.trend.test), "GEE", "GEEtrend", "GEEall" perform the GEE-based test (GEE.trend.test) with constant, linearly modeled, and freely varying scale parameters, respectively.
exact: logical, should an exact permutation test be performed. Only an exact test can be performed for "SO". The default is to use the asymptotic p-values except for "SO".
R: integer, number of permutations for the exact test
control: an optional list of control settings for the stochastic order ("SO") test, usually a call to soControl. See there for the names of the settable control values and their effect.

Author

Aniko Szabo, aszabo@mcw.edu

Examples

Run this code


data(shelltox)
trend.test(shelltox, test="RS")
set.seed(5724)
#R=50 is too low to get a good estimate of the p-value
trend.test(shelltox, test="RS", R=50, exact=TRUE)