data.generation.wrap: Data Generation wrap-up (short-cut name: dg.w)

In fadedsoul/cdml: CDML

Description

Data Generation wrap-up (short-cut name: dg.w)

Usage

data.generation.wrap(simu = 100, samples = 1000, covariates = 2, model,
  trimLowerBound = -4, trimUpperBound = 4,
  parametricCurveOption = "polynom", ncores = ncores, sd = 8,
  noise.sd = 1)

Arguments

simu	number of times one would run, default = 100.
samples	number of samples, default = 1000
covariates	dimension of covariates
model	choose from "CTE", "IV"
trimLowerBound	the lower bound of treatment trimming t, default = -4.
trimUpperBound	the lower bound of treatment trimming t, default = 4.
parametricCurveOption	choose from "linear", "polynom", "polynom2", "polynom3", "mixture".
ncores	number of cores we use
sd	t = ∑ x_i + ε, the standard error of ε, choose from 1, 2, 3, 5, 8, 10, 15
noise.sd	the standard error of response generation assuming the response model is based on a normal distribution

Value

list that contains two parts: $data contains generated data, $true contains the true underlying response and the density at grid on treatment

fadedsoul/cdml documentation built on May 14, 2019, 8:01 a.m.