p_adjustment_anova: Adjust p-values for multiple comparisons

In pmartR: Panomics Marketplace - Quality Control and Statistical Analysis for Panomics Data

Adjust p-values for multiple comparisons

Description

Depending upon the pval_adjust method selected, the supplied p_values are compared against an adjusted pval_thresh value or the provided means are used to compute new statistics, p-values are computed and compared against the provided pval_thresh. A data.frame that indicates which of the tests are significant, 1 if significant or 0 if insignificant. If means is also provided and the p-value is signficant then the direction of the change is indicated by the sign on 1, i.e., means<0 and p_value<pval_thresh will return -1, similarly for means>0.

Usage

p_adjustment_anova(
  p_values,
  diff_mean,
  t_stats,
  sizes,
  pval_adjust_multcomp,
  pval_adjust_fdr
)

Arguments

p_values	A matrix (or data.frame) of p-values to be adjusted.
diff_mean	A matrix (or data.frame) of groups means that are to be compared
t_stats	A matrix (or data.frame) of t-test statistics resulting from from standard procedures
sizes	A matrix (or data.frame) of group sizes
pval_adjust_multcomp	character vector specifying the type of multiple comparisons adjustment to implement. A NULL value corresponds to no adjustment. Valid options include: holm, bonferroni, dunnett, tukey or none.
pval_adjust_fdr	character vector specifying the type of FDR adjustment to implement. A NULL value corresponds to no adjustment. Valid options include: bonferroni, BH, BY, fdr, or none.

Value

a data frame with the following columns: group means, global G-test statistic and corresponding p-value

Author(s)

Bryan Stanfill

pmartR documentation built on Oct. 15, 2024, 1:06 a.m.