Barrientos_power: Barrientos power

In diff-priv-ht/nonpmRegPkg: Implementation of Functions for "Test of Tests"

Barrientos power

Description

Simulate a p-value computation for the test proposed by Barrientos et al. many times and compute the power. This is done for several choices of the truncation parameter, a. The design matrix is set prior to the simulations.

Usage

Barrientos_power(
  X,
  groups,
  M,
  a_vector,
  effect_size,
  alpha,
  epsilon,
  nsims,
  reps
)

Arguments

X	The design matrix.
groups	A vector with the group number of each observation.
M	The number of subsamples when computing p-value.
a_vector	Vector of truncation levels.
effect_size	The quotient of the parameter of interest (beta) and the standard deviation of the noise (sigma).
alpha	The significance level.
epsilon	The privacy parameter.
nsims	Number of simulations for the p-value computation.
reps	Number of simulations for the reference distribution computation.

Value

Will return a vector of powers of the same length as 'a_vector', where output corresponds to the power with that level of truncation.

diff-priv-ht/nonpmRegPkg documentation built on Feb. 6, 2023, 5:22 p.m.