logistic_power: Test of Tests Logistic Power

View source: R/logistic_implementation.R

logistic_powerR Documentation

Test of Tests Logistic Power

Description

Simulate a p-value computation for logistic regression many times and compute the power. The design matrix is regenerated for each simulation.

Usage

logistic_power(nsims, n, p, groups, epsilon, effect_size, M, theta_0, alpha)

Arguments

nsims

Number of simulations.

n

Number of observations (number of rows of the design matrix)

p

Number of parameters (number of columns of the design matrix). Includes the intercept.

groups

A vector with the group number of each observation.

epsilon

The privacy parameter.

effect_size

The quotient of the parameter of interest (beta) and the standard deviation of the noise (sigma).

M

The number of subsamples when computing p-value.

theta_0

The threshold.

alpha

The significance level.

Value

Will return a vector of p-values.


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