GenerateOneCenter: Generate one data center

In kaz-yos/distributed: Examine Privacy Preserving Data Analysis Methods in Simulated Distributed Data Network

Description

Generate one data center with information of n patients. The parameters are given as four raw vectors. This function is intentionally designed to be a low level function working with raw numeric vector parameters for maximum flexibility. In actual simulation runs, this function should be called by GenerateDistResNet, not directly.

Usage

GenerateOneCenter(n, AssignCovariates, alphas, betas, survParams)

Arguments

n	study site-specific sample size
AssignCovariates	covariate generation functions that takes n and p as the only arguments.
alphas	parameter vector for treatment model including c(alpha0, alphaX)
betas	parameter vector for outcome model shared among binary and survival outcome models including c(beta0, betaX, betaA, betaXA).
survParams	vector of two. The first element is the baseline hazard of events in the exponential event time outcome model (lambda). The second element is the baseline hazard of censoring in the exponential censoring time model (lambda_c).

Value

data frame with id, covariates, treatment (A) related variables, binary outcome (Y) related variables, and survival outcome (time, event) related variables.

Author(s)

Kazuki Yoshida

kaz-yos/distributed documentation built on May 27, 2019, 4:50 a.m.