Navigation
SEARCH

Implementation of Multiple Testing Procedures

From MultXpert

Jump to: navigation, search

This article provides information on software implementation of commonly used multiple testing procedures.

SAS macros

P-value-based nonparametric and semiparametric multiple testing procedures

PvalProc macro computes adjusted p-values for popular nonparametric procedures (Bonferroni, Holm, fixed-sequence and fallback procedures) and semiparametric procedures (Hommel and Hochberg procedures) in general hypothesis testing problems with equally or unequally weighted null hypotheses.

Chain macro computes adjusted p-values for the nonparametric chain procedure in general hypothesis testing problems (equally or unequally weighted null hypotheses).

PvalCI macro computes adjusted/simultaneous confidence intervals for popular nonparametric procedures (Bonferroni, Holm, fixed-sequence and fallback procedures) in general hypothesis testing problems with equally or unequally weighted null hypotheses.

Parametric multiple testing procedures

ParProc macro computes adjusted p-values for selected parametric procedures (single-step and step-down Dunnett procedures) in one-sided hypothesis testing problems with a balanced one-way layout and equally weighted null hypotheses.

ParCI macro computes adjusted/simultaneous confidence intervals for selected parametric procedures (single-step and step-down Dunnett procedures) in one-sided hypothesis testing problems with a balanced one-way layout and equally weighted null hypotheses.

Sample size calculations

SAS code for performing sample size calculations in clinical trials with co-primary endpoints. See Huque Dmitrienko DAgostino 2013 for more information.

R packages and functions

multXpert package includes multiple functions to implement commonly used multiple testing procedures, including computation of adjusted p-values and adjusted/simultaneous confidence intervals.