rss.t.test {generalRSS}R Documentation

RSS t-test for one-sample and two-sample problems

Description

The rss.t.test function performs one- and two-sample t-tests on ranked set sample data using t approximations, with methods described by Ahn et al. (2014).

Usage

rss.t.test(
  data1,
  data2 = NULL,
  alpha = 0.05,
  alternative = "two.sided",
  mu0 = 0,
  method
)

Arguments

data1

A numeric data frame of ranked set samples with columns rank for ranks and y for data values.

data2

An optional numeric data frame of ranked set samples with columns rank for ranks and y for data values.

alpha

A numeric value specifying the confidence level for the interval.

alternative

A character string specifying the alternative hypothesis. Must be one of "two.sided" (default), "greater", or "less".

mu0

A numeric value indicating the hypothesized value of the mean (for a one-sample problem) or the mean difference (for a two-sample problem).

method

A character string specifying the method used to approximate the t-distribution. Must be either "sample" or "naive".

Details

This function performs a t-test on ranked set sample data for both one-sample and two-sample mean problems, using t approximations. For a one-sample test, provide data1 as a data frame with rank and y columns. For a two-sample test, provide both data1 and data2 with equal set sizes. The method parameter allows for two options to approximate the t-distribution: "sample" and "naive" as introduced by Ahn et al. (2014). The function compute the t-statistic, confidence interval, degrees of freedom, and p-value based on the provided RSS data and specified parameters.

Value

RSS_mean

The RSS mean estimate (for a one-sample problem) or a vector of RSS mean estimates for each group (for a two-sample problem).

CI

The confidence interval for the population mean (for a one-sample problem) or for the mean difference (for a two-sample problem).

t

The t-statistic for the test.

df

The degrees of freedom for the test.

p.value

The p-value for the test.

References

S. Ahn, J. Lim, and X. Wang. (2014) The student’s t approximation to distributions of pivotal statistics from ranked set samples. Journal of the Korean Statistical Society, 43, 643–652.

See Also

rss.simulation: used for simulating Ranked Set Samples (RSS), which can serve as input.

rss.sampling: used for sampling Ranked Set Samples (RSS) from a population data set, providing input data.

Examples

## Balanced RSS with a set size 3 and equal sample sizes of 6 for each stratum,
## using imperfect ranking from a normal distribution with a mean of 0.
rss.data1=rss.simulation(H=3,nsamp=c(6,6,6),dist="normal", rho=0.8,delta=0)

## one-sample t-test using 'naive' method
rss.t.test(data1=rss.data1, data2=NULL, alpha=0.05,
alternative="two.sided", mu0=0, method="naive")

## one-sample t-test using 'sample' method
rss.t.test(data1=rss.data1, data2=NULL, alpha=0.05,
alternative="two.sided", mu0=0, method="sample")

## Unbalanced RSS with a set size 3 and different sample sizes of 6, 10, and 8 for each stratum,
## using imperfect ranking from a normal distribution with a mean of 0.
rss.data2<-rss.simulation(H=3,nsamp=c(6,8,10),dist="normal", rho=0.8,delta=0)

## two-sample t-test using 'naive' method
rss.t.test(data1=rss.data1, data2=rss.data2, alpha=0.05,
alternative="two.sided", mu0=0, method="naive")

## two-sample t-test using 'sample' method
rss.t.test(data1=rss.data1, data2=rss.data2, alpha=0.05,
alternative="two.sided", mu0=0, method="sample")

[Package generalRSS version 0.1.3 Index]