Power calculations for two sample t tests and confidence intervals

Usage

two_t_power(
  n = NULL,
  delta = NULL,
  sd = 1,
  sig.level = 0.05,
  power = 0.8,
  tol = .Machine$double.eps^0.25
)

two_t_power_equiv(
  n = NULL,
  equiv = NULL,
  sd = 1,
  conf.level = 0.95,
  power = 0.8,
  tol = .Machine$double.eps^0.25
)

two_t_power_me(
  n = NULL,
  me = NULL,
  sd = 1,
  conf.level = 0.95,
  power = 0.8,
  tol = .Machine$double.eps^0.25
)

Arguments

n: number of observations (per group)
delta: true difference between the parameter of interest and the null hypothesis
sd: standard deviation (within group)
sig.level: significance level (for delta)
power: power of the test, or of having the given margin of error or smaller
tol: numerical tolerance used in root finding, the default providing (at least) four significant digits.
equiv: equivalence bounds
conf.level: confidence level (for equivalence and margin of error)
me: margin of error of the confidence interval