Gamma approximated pvalues — gamma

A wrapper around the compute_p_values function from the permAprox package using a gamma control distribution.

Usage

gamma_pvalues(
  result_list,
  alternative = "greater",
  nullValue = 0,
  fit_thresh = 0.2
)

result_list: A list object that is the result of the multiple_comparisons function with save_perm_stats == TRUE.
alternative: A string indicating what the alternative hypothesis should be. Default is greater.
nullValue: Numeric or character. Specifies the value around which the null distribution is centered. If set to "mean" or "median", the per-row mean or median of perm_stats is used instead. This allows testing against a null hypothesis other than zero or centering based on the empirical distribution.
fit_thresh: Numeric. Threshold on empirical p-values below which parametric fitting is applied. Default: 0.2 (parametric approximation if the empirical p-value is smaller than 0.2).

Provides a data.frame with the updated the results.