optimize_power_ANOVA: Optimize the parameters of our test for ANOVA data

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

Optimize the parameters of our test for ANOVA data

Description

Function that finds the parameters, M and theta_0, of our test that yield the highest power for a given combination of n, the effect size, and epsilon

Usage

optimize_power_ANOVA(
  n,
  effect_size = 1,
  epsilon = 1,
  alpha = 0.05,
  groups = 3,
  M_max = floor(n/groups/2)
)

Arguments

n	The number of observations (number of rows in the database). Should be a multiple of three.
effect_size	The ratio of the between group variance to the within group variance. Defaults to 1, as in Couch et al.
epsilon	The privacy parameter. Defaults to 1, as in Couch et al.
alpha	The significance level. Defaults to 0.05.
groups	The number of groups in each subsample for ANOVA.
M_max	The maximum M to search over.

Value

The output will be a list with the achievable power and the corresponding two parameter values

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