mctp: Nonparametric multiple contrast tests and simultaneous confidence intervals (independent samples)

Description

The function mctp computes the estimator of nonparametric relative effects based on global rankings, simultaneous confidence intervals for the effects, and adjusted p-values based on contrasts in the setting of independent samples. Contrasts include "Tukey", "Dunnett", "Sequen", "Williams", "Changepoint", "AVE", "McDermott", "Marcus", "UmbrellaWilliams", "GrandMean", and "UserDefined". The statistics are computed using multivariate normal distribution, multivariate Satterthwaite t-Approximation, and multivariate transformations (adjusted log odds or Fisher function). The function 'mctp' computes both the one-sided and two-sided simultaneous confidence intervals and adjusted p-values. The simultaneous confidence intervals can be plotted.

Usage

mctp(formula, data, type = c("Tukey", "Dunnett", "Sequen", 
     "Williams", "Changepoint", "AVE", "McDermott", "Marcus", 
     "UmbrellaWilliams", "GrandMean", "UserDefined"), 
     conf.level = 0.95, alternative = c("two.sided", "less", 
     "greater"), asy.method = c("fisher", "mult.t", 
     "normal", "log.odds"), plot.simci = FALSE, control = NULL, info = TRUE, 
     rounds = 3, contrast.matrix = NULL, correlation = FALSE, 
     effect=c("unweighted","weighted"), const=1/1.702)

Arguments

formula

A two-sided 'formula' specifying a numeric response variable and a factor with more than two levels. If the factor contains less than 3 levels, an error message will be returned.

data

A dataframe containing the variables specified in formula.

type

Character string defining the type of contrast. It should be one of "Tukey", "Dunnett", "Sequen", "Williams", "Changepoint", "AVE", "McDermott", "Marcus", "UmbrellaWilliams", "GrandMean", "UserDefined".

conf.level

The confidence level for conf.level-confidence intervals (default is 0.95).

alternative

Character string defining the alternative hypothesis, one of "two.sided", "less", or "greater".

asy.method

Character string defining the asymptotic approximation method, one of "fisher" (for using the Fisher transformation function), "log.odds" (for using the adjusted log odds effect sizes), "mult.t" (for using a multivariate t-distribution with a Satterthwaite Approximation), , or "normal" (for using the multivariate normal distribution), "log.odds" (for using the adjusted log odds effect sizes).

plot.simci

A logical indicating whether you want a plot of the confidence intervals.

control

Character string defining the control group in Dunnett comparisons. By default, it is the first group by definition of the factor variable.

info

A logical whether you want a brief overview with informations about the output.

rounds

Number of rounds for the numeric values of the output (default is 3).

contrast.matrix

User-defined contrast matrix.

correlation

A logical whether the estimated correlation matrix and covariance matrix should be printed.

effect

Character string defining the type of effect, one of "unweighted" and "weighted".

const

Number used for the adjustment of log odds when the "log.odds" option is chosen.

Value

Data.Info

List of samples and sample sizes and estimated effect per group.

Contrast

Contrast matrix.

Analysis

Estimator: Estimated relative effect, Lower: Lower limit of the simultaneous confidence interval, Upper: Upper limit of the simultaneous confidence interval, Statistic: Test statistic p.Value: Adjusted p-values for the hypothesis by the choosen approximation method.

Analysis.Inf

The same as Analysis except that it assumes rounds = Inf.

Overall

The critical value and adjusted p-value for the overall hypothesis.

input

List of input arguments by user.

text.Output

Character string specifying the alternative hypotheses.

text.output.W

Character string specifying the weight pattern for the reference distribution.

connames

Character string specifying the contrast names.

AsyMethod

Character string specifying the approximation method.

References

F. Konietschke, L.A. Hothorn, E. Brunner: Rank-Based Multiple Test Procedures and Simultaneous Confidence Intervals. Electronic Journal of Statistics, Vol.0 (2011) 1-8.

Konietschke, F., Placzek, M., Schaarschmidt, S., Hothorn, L.A. (2015). nparcomp: An R Software Package for Nonparametric Multiple Comparisons and Simultaneous Confidence Intervals. Journal of Statistical Software, 61(10), 1-17.

Examples

Run this code

# NOT RUN {
data(liver)

  # Williams Contrast

a<-mctp(weight ~dosage, data=liver, asy.method = "fisher",
        type = "Williams", alternative = "two.sided", 
        plot.simci = TRUE, info = FALSE)
summary(a)

 # Dunnett Contrast

b<-mctp(weight ~dosage, data=liver, asy.method = "fisher",
        type = "Dunnett", alternative = "two.sided", 
        plot.simci = TRUE, info = FALSE)
summary(b)

 # Dunnett dose 3 is baseline

c<-mctp(weight ~dosage, data=liver, asy.method = "log.odds",
        type = "Dunnett", control = "3",alternative = "two.sided",
        plot.simci = TRUE, info = FALSE)
summary(c)


data(colu)

  # Tukey comparison- one sided (less)

a<-mctp(corpora~ dose, data=colu, asy.method = "log.odds",
        type = "Tukey",alternative = "less", 
        plot.simci = TRUE, info = FALSE)
summary(a)

 # Tukey comparison- one sided (greater)

b<-mctp(corpora~ dose, data=colu, asy.method = "mult.t",
        type = "Tukey",alternative = "greater", 
        plot.simci = TRUE, info = FALSE)
summary(b)

  # Tukey comparison- one sided (less)

c<-mctp(corpora~ dose, data=colu, asy.method = "mult.t",
        type = "Tukey",alternative = "less", 
        plot.simci = TRUE, info = FALSE)
summary(c)

 # Marcus comparison- one sided (greater)

d<-mctp(corpora~ dose, data=colu, asy.method = "fisher",
        type = "Marcus",alternative = "greater", 
        plot.simci = TRUE, info = FALSE)
summary(d)
# }