causalPAF (version 1.2.5)

do_sim_sequentialPAF: Simulates a Fitted Model for a Mediator or Exposure or Risk Factor Allowing for Potential Outcomes in Causal Analysis

Description

A fitted model for a mediator or exposure or risk factor can be simulated given values of the other risk factors or exposure saved in the data frame current_mat. This allows for potential outcomes to be measured for causal analysis. For example, for an outcome \(Y_{A,M}\) with exposure A and mediators \(M_{1}, M_{3}, \dots M_{K}\) the function can measure potential outcomes such as \(Y_{A=0,M_{1},M_{2},M_{3}}\) or \(Y_{A=0,M_{1},M_{2}=0,M_{3}=0}\) when there are three mediators. The model can be either a binary, continuous or an ordered factor response model.

Usage

do_sim_sequentialPAF(colnum, current_mat, model)

Value

simulation

simulation

Arguments

colnum

Column number of exposure or risk factor of interest within the data frame. The data frame has cases in rows and variables in columns.

current_mat

The data frame containing the data for which the model can be simulated with. For potential outcomes for example such as \(Y_{A=0,M_{1},M_{2},M_{3}}\) requires the exposure in this case to be pre set to zero i.e. current_mat should have the exposure \(Y_{A=0}\) set to zero if simulating e.g. \(M_{1}\).

model

A fitted causal regression model for either a binary, continuous or an ordered factor response.

Examples

if (FALSE) {
# I don't want you to run this
}