normal_confects: Confident effect sizes from from normal or t distributions

In topconfects: Top Confident Effect Sizes

Description

A general purpose confident effect size function for where a normal or t distribution of errors can be assumed. Calculates confident effect sizes based on an estimated effect and standard deviation (normal distribution), or mean and scale (t distribution).

Usage

normal_confects(
  effect,
  se,
  df = Inf,
  signed = TRUE,
  fdr = 0.05,
  step = 0.001,
  full = FALSE
)

Arguments

effect	A vector of estimated effects.
se	A single number or vector of standard errors (or if t distribution, scales).
df	A single number or vector of degrees of freedom, for t-distribution. Inf for normal distribution.
signed	If TRUE effects are signed, use TREAT test. If FALSE effects are all positive, use one sided t-test.
fdr	False Discovery Rate to control for.
step	Granularity of effect sizes to test.
full	Include some further statistics used to calculate confects in the output, and also include FDR-adjusted p-values that effect size is non-zero (note that this is against the spirit of the topconfects approach).

Value

See nest_confects for details of how to interpret the result.

Examples

# Find largest positive or negative z-scores in a collection,
# and place confidence bounds on them that maintain FDR 0.05.
z <- c(1,-2,3,-4,5)
normal_confects(z, se=1, fdr=0.05, full=TRUE)

topconfects documentation built on Nov. 8, 2020, 6:25 p.m.