multcomp (version 0.3-8)

csimint: Simultaneous Confidence Intervals Based on Parameter Estimates

Description

Computes simultaneous intervals for several multiple procedures based on precomputed parameter estimates.

Usage

csimint(estpar, df, covm, cmatrix = NULL, 
        ctype = "user-defined", conf.level = 0.95, 
        alternative = c("two.sided", "less", "greater"),
        asympt = FALSE, eps = 0.001, maxpts = 1e+06)

Arguments

estpar

estimated parameter vector.

df

degrees of freedom.

covm

estimated covariance matrix of estpar.

cmatrix

contrast matrix.

ctype

a string decribing the kind of contrast matrix used. Only used for printing in print.hmtest.

conf.level

confidence level.

alternative

the alternative hypothesis must be one of "two.sided" (default), "greater" or "less". You can specify just the initial letter.

asympt

a logical indicating whether the (exact) t-distribution or the normal approximation should be used.

eps

absolute error tolerance as double.

maxpts

maximum number of function values as integer.

Value

• an object of class hmtest

Details

This is the work-horse for simint. If only parameter estimates and estimates of the (co)variances are available, simultaneous confidence intervals can be computed with this low-level function. See cholesterol for an example.

References

Frank Bretz, Alan Genz and Ludwig A. Hothorn (2001), On the numerical availability of multiple comparison procedures. Biometrical Journal, 43(5), 645--656.