Nothing
R/bias_params.R
In multibias: Simultaneous Multi-Bias Adjustment
Defines functions
print.bias_params
bias_params
Documented in
bias_params
#' Represent bias parameters
#'
#' @description
#' `bias_params` is one of two different options to represent bias assumptions for bias adjustment. The [multibias_adjust()] function will apply the assumptions from these models and use them to adjust for biases in the observed data. It takes one input, a list, where each item in the list corresponds to the necessary models for bias adjustment. See below for bias models.
#'
#' For each of the following bias models, the variables are defined:
#' \itemize{
#' \item X = True exposure
#' \item X* = Misclassified exposure
#' \item Y = True outcome
#' \item Y* = Misclassified outcome
#' \item C = Known confounder
#' \item j = Number of known confounders
#' \item U = Uncontrolled confounder
#' \item S = Selection indicator
#' }
#'
#' \describe{
#' \item{Uncontrolled confounding}{
#' \ifelse{html}{\out{logit(P(U=1)) = α<sub>0</sub> + α<sub>1</sub>X + α<sub>2</sub>Y + α<sub>2+j</sub>C<sub>j</sub> }}{\eqn{logit(P(U=1)) = \alpha_0 + \alpha_1 X + \alpha_2 Y + \alpha_{2+j} C_j }}
#' }
#'
#' \item{Exposure misclassification}{
#' \ifelse{html}{\out{logit(P(X=1)) = δ<sub>0</sub> + δ<sub>1</sub>X* + δ<sub>2</sub>Y + δ<sub>2+j</sub>C<sub>j</sub> }}{\eqn{logit(P(X=1)) = \delta_0 + \delta_1 X^* + \delta_2 Y + \delta_{2+j} C_j }}
#' }
#'
#' \item{Outcome misclassification}{
#' \ifelse{html}{\out{logit(P(Y=1)) = δ<sub>0</sub> + δ<sub>1</sub>X + δ<sub>2</sub>Y* + δ<sub>2+j</sub>C<sub>j</sub> }}{\eqn{logit(P(Y=1)) = \delta_0 + \delta_1 X + \delta_2 Y^* + \delta_{2+j} C_j }}
#' }
#'
#' \item{Selection bias}{
#' \ifelse{html}{\out{logit(P(S=1)) = β<sub>0</sub> + β<sub>1</sub>X + β<sub>2</sub>Y }}{\eqn{logit(P(S=1)) = \beta_0 + \beta_1 X + \beta_2 Y }}
#' }
#'
#' \item{Uncontrolled Confounding & Exposure Misclassification (Option 1)}{
#' \ifelse{html}{\out{logit(P(U=1)) = α<sub>0</sub> + α<sub>1</sub>X + α<sub>2</sub>Y <br> logit(P(X=1)) = δ<sub>0</sub> + δ<sub>1</sub>X* + δ<sub>2</sub>Y + δ<sub>2+j</sub>C<sub>j</sub> }}{\eqn{logit(P(U=1)) = \alpha_0 + \alpha_1 X + \alpha_2 Y \\ logit(P(X=1)) = \delta_0 + \delta_1 X^* + \delta_2 Y + \delta_{2+j} C_j }}
#' }
#'
#' \item{Uncontrolled Confounding & Exposure Misclassification (Option 2)}{
#' \ifelse{html}{\out{log(P(X=1,U=0)/P(X=0,U=0)) = γ<sub>1,0</sub> + γ<sub>1,1</sub>X* + γ<sub>1,2</sub>Y + γ<sub>1,2+j</sub>C<sub>j</sub> <br> log(P(X=0,U=1)/P(X=0,U=0)) = γ<sub>2,0</sub> + γ<sub>2,1</sub>X* + γ<sub>2,2</sub>Y + γ<sub>2,2+j</sub>C<sub>j</sub> <br> log(P(X=1,U=1)/P(X=0,U=0)) = γ<sub>3,0</sub> + γ<sub>3,1</sub>X* + γ<sub>3,2</sub>Y + γ<sub>3,2+j</sub>C<sub>j</sub> }}{\eqn{log(P(X=1,U=0)/P(X=0,U=0)) = \gamma_{1,0} + \gamma_{1,1} X^* + \gamma_{1,2} Y + \gamma_{1,2+j} C_j \\ log(P(X=0,U=1)/P(X=0,U=0)) = \gamma_{2,0} + \gamma_{2,1} X^* + \gamma_{2,2} Y + \gamma_{2,2+j} C_j \\ log(P(X=1,U=1)/P(X=0,U=0)) = \gamma_{3,0} + \gamma_{3,1} X^* + \gamma_{3,2} Y + \gamma_{3,2+j} C_j }}
#' }
#'
#' \item{Uncontrolled Confounding & Outcome Misclassification (Option 1)}{
#' \ifelse{html}{\out{logit(P(U=1)) = α<sub>0</sub> + α<sub>1</sub>X + α<sub>2</sub>Y <br> logit(P(Y=1)) = δ<sub>0</sub> + δ<sub>1</sub>X + δ<sub>2</sub>Y* + δ<sub>2+j</sub>C<sub>j</sub> }}{\eqn{logit(P(U=1)) = \alpha_0 + \alpha_1 X + \alpha_2 Y \\ logit(P(Y=1)) = \delta_0 + \delta_1 X + \delta_2 Y^* + \delta_{2+j} C_j }}
#' }
#'
#' \item{Uncontrolled Confounding & Outcome Misclassification (Option 2)}{
#' \ifelse{html}{\out{log(P(U=1,Y=0)/P(U=0,Y=0)) = γ<sub>1,0</sub> + γ<sub>1,1</sub>X + γ<sub>1,2</sub>Y* + γ<sub>1,2+j</sub>C<sub>j</sub> <br> log(P(U=0,Y=1)/P(U=0,Y=0)) = γ<sub>2,0</sub> + γ<sub>2,1</sub>X + γ<sub>2,2</sub>Y* + γ<sub>2,2+j</sub>C<sub>j</sub> <br> log(P(U=1,Y=1)/P(U=0,Y=0)) = γ<sub>3,0</sub> + γ<sub>3,1</sub>X + γ<sub>3,2</sub>Y* + γ<sub>3,2+j</sub>C<sub>j</sub> }}{\eqn{log(P(U=1,Y=0)/P(U=0,Y=0)) = \gamma_{1,0} + \gamma_{1,1} X + \gamma_{1,2} Y^* + \gamma_{1,2+j} C_j \\ log(P(U=0,Y=1)/P(U=0,Y=0)) = \gamma_{2,0} + \gamma_{2,1} X + \gamma_{2,2} Y^* + \gamma_{2,2+j} C_j \\ log(P(U=1,Y=1)/P(U=0,Y=0)) = \gamma_{3,0} + \gamma_{3,1} X + \gamma_{3,2} Y^* + \gamma_{3,2+j} C_j }}
#' }
#'
#' \item{Uncontrolled Confounding & Selection Bias}{
#' \ifelse{html}{\out{logit(P(U=1)) = α<sub>0</sub> + α<sub>1</sub>X + α<sub>2</sub>Y + α<sub>2+j</sub>C<sub>j</sub> <br> logit(P(S=1)) = β<sub>0</sub> + β<sub>1</sub>X + β<sub>2</sub>Y }}{\eqn{logit(P(U=1)) = \alpha_0 + \alpha_1 X + \alpha_2 Y + \alpha_{2+j} C_j \\ logit(P(S=1)) = \beta_0 + \beta_1 X + \beta_2 Y }}
#' }
#'
#' \item{Exposure Misclassification & Outcome Misclassification (Option 1)}{
#' \ifelse{html}{\out{logit(P(X=1)) = δ<sub>0</sub> + δ<sub>1</sub>X* + δ<sub>2</sub>Y* + δ<sub>2+j</sub>C<sub>j</sub> <br> logit(P(Y=1)) = β<sub>0</sub> + β<sub>1</sub>X + β<sub>2</sub>Y* + β<sub>2+j</sub>C<sub>j</sub> }}{\eqn{logit(P(X=1)) = \delta_0 + \delta_1 X^* + \delta_2 Y^* + \delta_{2+j} C_j \\ logit(P(Y=1)) = \beta_0 + \beta_1 X + \beta_2 Y^* + \beta_{2+j} C_j }}
#' }
#'
#' \item{Exposure Misclassification & Outcome Misclassification (Option 2)}{
#' \ifelse{html}{\out{log(P(X=1,Y=0) / P(X=0,Y=0)) = γ<sub>1,0</sub> + γ<sub>1,1</sub>X* + γ<sub>1,2</sub>Y* + γ<sub>1,2+j</sub>C<sub>j</sub> <br> log(P(X=0,Y=1) / P(X=0,Y=0)) = γ<sub>2,0</sub> + γ<sub>2,1</sub>X* + γ<sub>2,2</sub>Y* + γ<sub>2,2+j</sub>C<sub>j</sub> <br> log(P(X=1,Y=1) / P(X=0,Y=0)) = γ<sub>3,0</sub> + γ<sub>3,1</sub>X* + γ<sub>3,2</sub>Y* + γ<sub>3,2+j</sub>C<sub>j</sub> }}{\eqn{log(P(X=1,Y=0) / P(X=0,Y=0)) = \gamma_{1,0} + \gamma_{1,1} X^* + \gamma_{1,2} Y^* + \gamma_{1,2+j} C_j \\ log(P(X=0,Y=1) / P(X=0,Y=0)) = \gamma_{2,0} + \gamma_{2,1} X^* + \gamma_{2,2} Y^* + \gamma_{2,2+j} C_j \\ log(P(X=1,Y=1) / P(X=0,Y=0)) = \gamma_{3,0} + \gamma_{3,1} X^* + \gamma_{3,2} Y^* + \gamma_{3,2+j} C_j }}
#' }
#'
#' \item{Exposure Misclassification & Selection Bias}{
#' \ifelse{html}{\out{logit(P(X=1)) = δ<sub>0</sub> + δ<sub>1</sub>X* + δ<sub>2</sub>Y + δ<sub>2+j</sub>C<sub>j</sub> <br> logit(P(S=1)) = β<sub>0</sub> + β<sub>1</sub>X* + β<sub>2</sub>Y + β<sub>2+j</sub>C<sub>j</sub> }}{\eqn{logit(P(X=1)) = \delta_0 + \delta_1 X^* + \delta_2 Y + \delta_{2+j} C_j \\ logit(P(S=1)) = \beta_0 + \beta_1 X^* + \beta_2 Y + \beta_{2+j} C_j }}
#' }
#'
#' \item{Outcome Misclassification & Selection Bias}{
#' \ifelse{html}{\out{logit(P(Y=1)) = δ<sub>0</sub> + δ<sub>1</sub>X + δ<sub>2</sub>Y* + δ<sub>2+j</sub>C<sub>j</sub> <br> logit(P(S=1)) = β<sub>0</sub> + β<sub>1</sub>X + β<sub>2</sub>Y* + β<sub>2+j</sub>C<sub>j</sub> }}{\eqn{logit(P(Y=1)) = \delta_0 + \delta_1 X + \delta_2 Y^* + \delta_{2+j} C_j \\ logit(P(S=1)) = \beta_0 + \beta_1 X + \beta_2 Y^* + \beta_{2+j} C_j }}
#' }
#' \item{Uncontrolled Confounding, Exposure Misclassification, and Selection Bias (Option 1)}{
#' \ifelse{html}{\out{logit(P(U=1)) = α<sub>0</sub> + α<sub>1</sub>X + α<sub>2</sub>Y <br> logit(P(X=1)) = δ<sub>0</sub> + δ<sub>1</sub>X* + δ<sub>2</sub>Y + δ<sub>2+j</sub>C<sub>j</sub> <br> logit(P(S=1)) = β<sub>0</sub> + β<sub>1</sub>X* + β<sub>2</sub>Y + β<sub>2+j</sub>C<sub>j</sub> }}{\eqn{logit(P(U=1)) = \alpha_0 + \alpha_1 X + \alpha_2 Y \\ logit(P(X=1)) = \delta_0 + \delta_1 X^* + \delta_2 Y + \delta_{2+j} C_j \\ logit(P(S=1)) = \beta_0 + \beta_1 X^* + \beta_2 Y + \beta_{2+j} C_j }}
#' }
#' \item{Uncontrolled Confounding, Exposure Misclassification, and Selection Bias (Option 2)}{
#' \ifelse{html}{\out{log(P(X=1,U=0)/P(X=0,U=0)) = γ<sub>1,0</sub> + γ<sub>1,1</sub>X* + γ<sub>1,2</sub>Y + γ<sub>1,2+j</sub>C<sub>j</sub> <br> log(P(X=0,U=1)/P(X=0,U=0)) = γ<sub>2,0</sub> + γ<sub>2,1</sub>X* + γ<sub>2,2</sub>Y + γ<sub>2,2+j</sub>C<sub>j</sub> <br> log(P(X=1,U=1)/P(X=0,U=0)) = γ<sub>3,0</sub> + γ<sub>3,1</sub>X* + γ<sub>3,2</sub>Y + γ<sub>3,2+j</sub>C<sub>j</sub> <br> logit(P(S=1)) = β<sub>0</sub> + β<sub>1</sub>X* + β<sub>2</sub>Y + β<sub>2+j</sub>C<sub>j</sub> }}{\eqn{log(P(X=1,U=0)/P(X=0,U=0)) = \gamma_{1,0} + \gamma_{1,1} X^* + \gamma_{1,2} Y + \gamma_{1,2+j} C_j \\ log(P(X=0,U=1)/P(X=0,U=0)) = \gamma_{2,0} + \gamma_{2,1} X^* + \gamma_{2,2} Y + \gamma_{2,2+j} C_j \\ log(P(X=1,U=1)/P(X=0,U=0)) = \gamma_{3,0} + \gamma_{3,1} X^* + \gamma_{3,2} Y + \gamma_{3,2+j} C_j \\ logit(P(S=1)) = \beta_0 + \beta_1 X^* + \beta_2 Y + \beta_{2+j} C_j }}
#' }
#' \item{Uncontrolled Confounding, Outcome Misclassification, and Selection Bias (Option 1)}{
#' \ifelse{html}{\out{logit(P(U=1)) = α<sub>0</sub> + α<sub>1</sub>X + α<sub>2</sub>Y <br> logit(P(Y=1)) = δ<sub>0</sub> + δ<sub>1</sub>X + δ<sub>2</sub>Y* + δ<sub>2+j</sub>C<sub>j</sub> <br> logit(P(S=1)) = β<sub>0</sub> + β<sub>1</sub>X + β<sub>2</sub>Y* + β<sub>2+j</sub>C<sub>j</sub> }}{\eqn{logit(P(U=1)) = \alpha_0 + \alpha_1 X + \alpha_2 Y \\ logit(P(Y=1)) = \delta_0 + \delta_1 X + \delta_2 Y^* + \delta_{2+j} C_j \\ logit(P(S=1)) = \beta_0 + \beta_1 X + \beta_2 Y^* + \beta_{2+j} C_j }}
#' }
#' \item{Uncontrolled Confounding, Outcome Misclassification, and Selection Bias (Option 2)}{
#' \ifelse{html}{\out{log(P(U=1,Y=0)/P(U=0,Y=0)) = γ<sub>1,0</sub> + γ<sub>1,1</sub>X + γ<sub>1,2</sub>Y* + γ<sub>1,2+j</sub>C<sub>j</sub> <br> log(P(U=0,Y=1)/P(U=0,Y=0)) = γ<sub>2,0</sub> + γ<sub>2,1</sub>X + γ<sub>2,2</sub>Y* + γ<sub>2,2+j</sub>C<sub>j</sub> <br> log(P(U=1,Y=1)/P(U=0,Y=0)) = γ<sub>3,0</sub> + γ<sub>3,1</sub>X + γ<sub>3,2</sub>Y* + γ<sub>3,2+j</sub>C<sub>j</sub> <br> logit(P(S=1)) = β<sub>0</sub> + β<sub>1</sub>X + β<sub>2</sub>Y* + β<sub>2+j</sub>C<sub>j</sub> }}{\eqn{log(P(U=1,Y=0)/P(U=0,Y=0)) = \gamma_{1,0} + \gamma_{1,1} X + \gamma_{1,2} Y^* + \gamma_{1,2+j} C_j \\ log(P(U=0,Y=1)/P(U=0,Y=0)) = \gamma_{2,0} + \gamma_{2,1} X + \gamma_{2,2} Y^* + \gamma_{2,2+j} C_j \\ log(P(U=1,Y=1)/P(U=0,Y=0)) = \gamma_{3,0} + \gamma_{3,1} X + \gamma_{3,2} Y^* + \gamma_{3,2+j} C_j \\ logit(P(S=1)) = \beta_0 + \beta_1 X + \beta_2 Y^* + \beta_{2+j} C_j }}
#' }
#' }
#'
#' @param coef_list List of coefficient values from the above options of
#' models. Each item of the list is an equation. The left side of the equation
#' identifies the model (i.e., "u" for the model predicting the uncontrolled
#' confounder). For the multinomial models, specify the value here based on the
#' numerator (i.e., "x1u0", "x0u1", "x1u1" for the three multinomial models in
#' Uncontrolled Confounding & Exposure Misclassification, Option 2)
#' The right side of the equation is the vector of values
#' corresponding to the model coefficients (from left to right).
#'
#' @examples
#' list_for_uc <- list(
#' u = c(-0.19, 0.61, 0.70, -0.09, 0.10, -0.15)
#' )
#'
#' bp_uc <- bias_params(coef_list = list_for_uc)
#'
#' list_for_em_om <- list(
#' x1y0 = c(-2.18, 1.63, 0.23, 0.36),
#' x0y1 = c(-3.17, 0.22, 1.60, 0.40),
#' x1y1 = c(-4.76, 1.82, 1.83, 0.72)
#' )
#'
#' bp_em_om <- bias_params(coef_list = list_for_em_om)
#'
#' @export
bias_params <- function(coef_list) {
# Check if the input is a list (R's equivalent of a dictionary)
if (!is.list(coef_list)) {
stop("Input must be a list (dictionary).")
}
# Check if all keys are strings
if (!all(sapply(names(coef_list), is.character))) {
stop("All keys in the dictionary must be strings.")
}
# Check if all values are vectors
if (!all(sapply(coef_list, is.vector))) {
stop("All values in the dictionary must be vectors.")
}
# Check if key is an acceptable value
acceptable_values <- c(
"u", "x", "y", "s", "x1u0", "x0u1", "x1u1", "u1y0", "u0y1", "u1y1",
"x1y0", "x0y1", "x1y1"
)
for (name in names(coef_list)) {
if (!(name %in% acceptable_values)) {
stop(
paste0(
"Unacceptable value in coef_list. Acceptable values include: ",
"u, x, y, s, x1u0, x1u1, x1u1, u1y0, u0y1, u1y1, x1y0, x0y1, x1y1"
)
)
}
}
obj <- list(coef_list = coef_list)
class(obj) <- "bias_params"
return(obj)
}
#' @export
print.bias_params <- function(x, ...) {
cat("Bias Parameters\n")
cat("---------------------------------\n")
for (key in names(x$coef_list)) {
cat("Model", key, "\n")
cat(" Coefs: ", paste(x$coef_list[[key]], collapse = ", "), "\n")
}
invisible(x)
}
Try the multibias package in your browser
Any scripts or data that you put into this service are public.
multibias documentation built on April 11, 2025, 5:53 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Defines functions print.bias_params bias_params
Documented in bias_params
#' Represent bias parameters
#'
#' @description
#' `bias_params` is one of two different options to represent bias assumptions for bias adjustment. The [multibias_adjust()] function will apply the assumptions from these models and use them to adjust for biases in the observed data. It takes one input, a list, where each item in the list corresponds to the necessary models for bias adjustment. See below for bias models.
#'
#' For each of the following bias models, the variables are defined:
#' \itemize{
#' \item X = True exposure
#' \item X* = Misclassified exposure
#' \item Y = True outcome
#' \item Y* = Misclassified outcome
#' \item C = Known confounder
#' \item j = Number of known confounders
#' \item U = Uncontrolled confounder
#' \item S = Selection indicator
#' }
#'
#' \describe{
#' \item{Uncontrolled confounding}{
#' \ifelse{html}{\out{logit(P(U=1)) = α<sub>0</sub> + α<sub>1</sub>X + α<sub>2</sub>Y + α<sub>2+j</sub>C<sub>j</sub> }}{\eqn{logit(P(U=1)) = \alpha_0 + \alpha_1 X + \alpha_2 Y + \alpha_{2+j} C_j }}
#' }
#'
#' \item{Exposure misclassification}{
#' \ifelse{html}{\out{logit(P(X=1)) = δ<sub>0</sub> + δ<sub>1</sub>X* + δ<sub>2</sub>Y + δ<sub>2+j</sub>C<sub>j</sub> }}{\eqn{logit(P(X=1)) = \delta_0 + \delta_1 X^* + \delta_2 Y + \delta_{2+j} C_j }}
#' }
#'
#' \item{Outcome misclassification}{
#' \ifelse{html}{\out{logit(P(Y=1)) = δ<sub>0</sub> + δ<sub>1</sub>X + δ<sub>2</sub>Y* + δ<sub>2+j</sub>C<sub>j</sub> }}{\eqn{logit(P(Y=1)) = \delta_0 + \delta_1 X + \delta_2 Y^* + \delta_{2+j} C_j }}
#' }
#'
#' \item{Selection bias}{
#' \ifelse{html}{\out{logit(P(S=1)) = β<sub>0</sub> + β<sub>1</sub>X + β<sub>2</sub>Y }}{\eqn{logit(P(S=1)) = \beta_0 + \beta_1 X + \beta_2 Y }}
#' }
#'
#' \item{Uncontrolled Confounding & Exposure Misclassification (Option 1)}{
#' \ifelse{html}{\out{logit(P(U=1)) = α<sub>0</sub> + α<sub>1</sub>X + α<sub>2</sub>Y <br> logit(P(X=1)) = δ<sub>0</sub> + δ<sub>1</sub>X* + δ<sub>2</sub>Y + δ<sub>2+j</sub>C<sub>j</sub> }}{\eqn{logit(P(U=1)) = \alpha_0 + \alpha_1 X + \alpha_2 Y \\ logit(P(X=1)) = \delta_0 + \delta_1 X^* + \delta_2 Y + \delta_{2+j} C_j }}
#' }
#'
#' \item{Uncontrolled Confounding & Exposure Misclassification (Option 2)}{
#' \ifelse{html}{\out{log(P(X=1,U=0)/P(X=0,U=0)) = γ<sub>1,0</sub> + γ<sub>1,1</sub>X* + γ<sub>1,2</sub>Y + γ<sub>1,2+j</sub>C<sub>j</sub> <br> log(P(X=0,U=1)/P(X=0,U=0)) = γ<sub>2,0</sub> + γ<sub>2,1</sub>X* + γ<sub>2,2</sub>Y + γ<sub>2,2+j</sub>C<sub>j</sub> <br> log(P(X=1,U=1)/P(X=0,U=0)) = γ<sub>3,0</sub> + γ<sub>3,1</sub>X* + γ<sub>3,2</sub>Y + γ<sub>3,2+j</sub>C<sub>j</sub> }}{\eqn{log(P(X=1,U=0)/P(X=0,U=0)) = \gamma_{1,0} + \gamma_{1,1} X^* + \gamma_{1,2} Y + \gamma_{1,2+j} C_j \\ log(P(X=0,U=1)/P(X=0,U=0)) = \gamma_{2,0} + \gamma_{2,1} X^* + \gamma_{2,2} Y + \gamma_{2,2+j} C_j \\ log(P(X=1,U=1)/P(X=0,U=0)) = \gamma_{3,0} + \gamma_{3,1} X^* + \gamma_{3,2} Y + \gamma_{3,2+j} C_j }}
#' }
#'
#' \item{Uncontrolled Confounding & Outcome Misclassification (Option 1)}{
#' \ifelse{html}{\out{logit(P(U=1)) = α<sub>0</sub> + α<sub>1</sub>X + α<sub>2</sub>Y <br> logit(P(Y=1)) = δ<sub>0</sub> + δ<sub>1</sub>X + δ<sub>2</sub>Y* + δ<sub>2+j</sub>C<sub>j</sub> }}{\eqn{logit(P(U=1)) = \alpha_0 + \alpha_1 X + \alpha_2 Y \\ logit(P(Y=1)) = \delta_0 + \delta_1 X + \delta_2 Y^* + \delta_{2+j} C_j }}
#' }
#'
#' \item{Uncontrolled Confounding & Outcome Misclassification (Option 2)}{
#' \ifelse{html}{\out{log(P(U=1,Y=0)/P(U=0,Y=0)) = γ<sub>1,0</sub> + γ<sub>1,1</sub>X + γ<sub>1,2</sub>Y* + γ<sub>1,2+j</sub>C<sub>j</sub> <br> log(P(U=0,Y=1)/P(U=0,Y=0)) = γ<sub>2,0</sub> + γ<sub>2,1</sub>X + γ<sub>2,2</sub>Y* + γ<sub>2,2+j</sub>C<sub>j</sub> <br> log(P(U=1,Y=1)/P(U=0,Y=0)) = γ<sub>3,0</sub> + γ<sub>3,1</sub>X + γ<sub>3,2</sub>Y* + γ<sub>3,2+j</sub>C<sub>j</sub> }}{\eqn{log(P(U=1,Y=0)/P(U=0,Y=0)) = \gamma_{1,0} + \gamma_{1,1} X + \gamma_{1,2} Y^* + \gamma_{1,2+j} C_j \\ log(P(U=0,Y=1)/P(U=0,Y=0)) = \gamma_{2,0} + \gamma_{2,1} X + \gamma_{2,2} Y^* + \gamma_{2,2+j} C_j \\ log(P(U=1,Y=1)/P(U=0,Y=0)) = \gamma_{3,0} + \gamma_{3,1} X + \gamma_{3,2} Y^* + \gamma_{3,2+j} C_j }}
#' }
#'
#' \item{Uncontrolled Confounding & Selection Bias}{
#' \ifelse{html}{\out{logit(P(U=1)) = α<sub>0</sub> + α<sub>1</sub>X + α<sub>2</sub>Y + α<sub>2+j</sub>C<sub>j</sub> <br> logit(P(S=1)) = β<sub>0</sub> + β<sub>1</sub>X + β<sub>2</sub>Y }}{\eqn{logit(P(U=1)) = \alpha_0 + \alpha_1 X + \alpha_2 Y + \alpha_{2+j} C_j \\ logit(P(S=1)) = \beta_0 + \beta_1 X + \beta_2 Y }}
#' }
#'
#' \item{Exposure Misclassification & Outcome Misclassification (Option 1)}{
#' \ifelse{html}{\out{logit(P(X=1)) = δ<sub>0</sub> + δ<sub>1</sub>X* + δ<sub>2</sub>Y* + δ<sub>2+j</sub>C<sub>j</sub> <br> logit(P(Y=1)) = β<sub>0</sub> + β<sub>1</sub>X + β<sub>2</sub>Y* + β<sub>2+j</sub>C<sub>j</sub> }}{\eqn{logit(P(X=1)) = \delta_0 + \delta_1 X^* + \delta_2 Y^* + \delta_{2+j} C_j \\ logit(P(Y=1)) = \beta_0 + \beta_1 X + \beta_2 Y^* + \beta_{2+j} C_j }}
#' }
#'
#' \item{Exposure Misclassification & Outcome Misclassification (Option 2)}{
#' \ifelse{html}{\out{log(P(X=1,Y=0) / P(X=0,Y=0)) = γ<sub>1,0</sub> + γ<sub>1,1</sub>X* + γ<sub>1,2</sub>Y* + γ<sub>1,2+j</sub>C<sub>j</sub> <br> log(P(X=0,Y=1) / P(X=0,Y=0)) = γ<sub>2,0</sub> + γ<sub>2,1</sub>X* + γ<sub>2,2</sub>Y* + γ<sub>2,2+j</sub>C<sub>j</sub> <br> log(P(X=1,Y=1) / P(X=0,Y=0)) = γ<sub>3,0</sub> + γ<sub>3,1</sub>X* + γ<sub>3,2</sub>Y* + γ<sub>3,2+j</sub>C<sub>j</sub> }}{\eqn{log(P(X=1,Y=0) / P(X=0,Y=0)) = \gamma_{1,0} + \gamma_{1,1} X^* + \gamma_{1,2} Y^* + \gamma_{1,2+j} C_j \\ log(P(X=0,Y=1) / P(X=0,Y=0)) = \gamma_{2,0} + \gamma_{2,1} X^* + \gamma_{2,2} Y^* + \gamma_{2,2+j} C_j \\ log(P(X=1,Y=1) / P(X=0,Y=0)) = \gamma_{3,0} + \gamma_{3,1} X^* + \gamma_{3,2} Y^* + \gamma_{3,2+j} C_j }}
#' }
#'
#' \item{Exposure Misclassification & Selection Bias}{
#' \ifelse{html}{\out{logit(P(X=1)) = δ<sub>0</sub> + δ<sub>1</sub>X* + δ<sub>2</sub>Y + δ<sub>2+j</sub>C<sub>j</sub> <br> logit(P(S=1)) = β<sub>0</sub> + β<sub>1</sub>X* + β<sub>2</sub>Y + β<sub>2+j</sub>C<sub>j</sub> }}{\eqn{logit(P(X=1)) = \delta_0 + \delta_1 X^* + \delta_2 Y + \delta_{2+j} C_j \\ logit(P(S=1)) = \beta_0 + \beta_1 X^* + \beta_2 Y + \beta_{2+j} C_j }}
#' }
#'
#' \item{Outcome Misclassification & Selection Bias}{
#' \ifelse{html}{\out{logit(P(Y=1)) = δ<sub>0</sub> + δ<sub>1</sub>X + δ<sub>2</sub>Y* + δ<sub>2+j</sub>C<sub>j</sub> <br> logit(P(S=1)) = β<sub>0</sub> + β<sub>1</sub>X + β<sub>2</sub>Y* + β<sub>2+j</sub>C<sub>j</sub> }}{\eqn{logit(P(Y=1)) = \delta_0 + \delta_1 X + \delta_2 Y^* + \delta_{2+j} C_j \\ logit(P(S=1)) = \beta_0 + \beta_1 X + \beta_2 Y^* + \beta_{2+j} C_j }}
#' }
#' \item{Uncontrolled Confounding, Exposure Misclassification, and Selection Bias (Option 1)}{
#' \ifelse{html}{\out{logit(P(U=1)) = α<sub>0</sub> + α<sub>1</sub>X + α<sub>2</sub>Y <br> logit(P(X=1)) = δ<sub>0</sub> + δ<sub>1</sub>X* + δ<sub>2</sub>Y + δ<sub>2+j</sub>C<sub>j</sub> <br> logit(P(S=1)) = β<sub>0</sub> + β<sub>1</sub>X* + β<sub>2</sub>Y + β<sub>2+j</sub>C<sub>j</sub> }}{\eqn{logit(P(U=1)) = \alpha_0 + \alpha_1 X + \alpha_2 Y \\ logit(P(X=1)) = \delta_0 + \delta_1 X^* + \delta_2 Y + \delta_{2+j} C_j \\ logit(P(S=1)) = \beta_0 + \beta_1 X^* + \beta_2 Y + \beta_{2+j} C_j }}
#' }
#' \item{Uncontrolled Confounding, Exposure Misclassification, and Selection Bias (Option 2)}{
#' \ifelse{html}{\out{log(P(X=1,U=0)/P(X=0,U=0)) = γ<sub>1,0</sub> + γ<sub>1,1</sub>X* + γ<sub>1,2</sub>Y + γ<sub>1,2+j</sub>C<sub>j</sub> <br> log(P(X=0,U=1)/P(X=0,U=0)) = γ<sub>2,0</sub> + γ<sub>2,1</sub>X* + γ<sub>2,2</sub>Y + γ<sub>2,2+j</sub>C<sub>j</sub> <br> log(P(X=1,U=1)/P(X=0,U=0)) = γ<sub>3,0</sub> + γ<sub>3,1</sub>X* + γ<sub>3,2</sub>Y + γ<sub>3,2+j</sub>C<sub>j</sub> <br> logit(P(S=1)) = β<sub>0</sub> + β<sub>1</sub>X* + β<sub>2</sub>Y + β<sub>2+j</sub>C<sub>j</sub> }}{\eqn{log(P(X=1,U=0)/P(X=0,U=0)) = \gamma_{1,0} + \gamma_{1,1} X^* + \gamma_{1,2} Y + \gamma_{1,2+j} C_j \\ log(P(X=0,U=1)/P(X=0,U=0)) = \gamma_{2,0} + \gamma_{2,1} X^* + \gamma_{2,2} Y + \gamma_{2,2+j} C_j \\ log(P(X=1,U=1)/P(X=0,U=0)) = \gamma_{3,0} + \gamma_{3,1} X^* + \gamma_{3,2} Y + \gamma_{3,2+j} C_j \\ logit(P(S=1)) = \beta_0 + \beta_1 X^* + \beta_2 Y + \beta_{2+j} C_j }}
#' }
#' \item{Uncontrolled Confounding, Outcome Misclassification, and Selection Bias (Option 1)}{
#' \ifelse{html}{\out{logit(P(U=1)) = α<sub>0</sub> + α<sub>1</sub>X + α<sub>2</sub>Y <br> logit(P(Y=1)) = δ<sub>0</sub> + δ<sub>1</sub>X + δ<sub>2</sub>Y* + δ<sub>2+j</sub>C<sub>j</sub> <br> logit(P(S=1)) = β<sub>0</sub> + β<sub>1</sub>X + β<sub>2</sub>Y* + β<sub>2+j</sub>C<sub>j</sub> }}{\eqn{logit(P(U=1)) = \alpha_0 + \alpha_1 X + \alpha_2 Y \\ logit(P(Y=1)) = \delta_0 + \delta_1 X + \delta_2 Y^* + \delta_{2+j} C_j \\ logit(P(S=1)) = \beta_0 + \beta_1 X + \beta_2 Y^* + \beta_{2+j} C_j }}
#' }
#' \item{Uncontrolled Confounding, Outcome Misclassification, and Selection Bias (Option 2)}{
#' \ifelse{html}{\out{log(P(U=1,Y=0)/P(U=0,Y=0)) = γ<sub>1,0</sub> + γ<sub>1,1</sub>X + γ<sub>1,2</sub>Y* + γ<sub>1,2+j</sub>C<sub>j</sub> <br> log(P(U=0,Y=1)/P(U=0,Y=0)) = γ<sub>2,0</sub> + γ<sub>2,1</sub>X + γ<sub>2,2</sub>Y* + γ<sub>2,2+j</sub>C<sub>j</sub> <br> log(P(U=1,Y=1)/P(U=0,Y=0)) = γ<sub>3,0</sub> + γ<sub>3,1</sub>X + γ<sub>3,2</sub>Y* + γ<sub>3,2+j</sub>C<sub>j</sub> <br> logit(P(S=1)) = β<sub>0</sub> + β<sub>1</sub>X + β<sub>2</sub>Y* + β<sub>2+j</sub>C<sub>j</sub> }}{\eqn{log(P(U=1,Y=0)/P(U=0,Y=0)) = \gamma_{1,0} + \gamma_{1,1} X + \gamma_{1,2} Y^* + \gamma_{1,2+j} C_j \\ log(P(U=0,Y=1)/P(U=0,Y=0)) = \gamma_{2,0} + \gamma_{2,1} X + \gamma_{2,2} Y^* + \gamma_{2,2+j} C_j \\ log(P(U=1,Y=1)/P(U=0,Y=0)) = \gamma_{3,0} + \gamma_{3,1} X + \gamma_{3,2} Y^* + \gamma_{3,2+j} C_j \\ logit(P(S=1)) = \beta_0 + \beta_1 X + \beta_2 Y^* + \beta_{2+j} C_j }}
#' }
#' }
#'
#' @param coef_list List of coefficient values from the above options of
#' models. Each item of the list is an equation. The left side of the equation
#' identifies the model (i.e., "u" for the model predicting the uncontrolled
#' confounder). For the multinomial models, specify the value here based on the
#' numerator (i.e., "x1u0", "x0u1", "x1u1" for the three multinomial models in
#' Uncontrolled Confounding & Exposure Misclassification, Option 2)
#' The right side of the equation is the vector of values
#' corresponding to the model coefficients (from left to right).
#'
#' @examples
#' list_for_uc <- list(
#' u = c(-0.19, 0.61, 0.70, -0.09, 0.10, -0.15)
#' )
#'
#' bp_uc <- bias_params(coef_list = list_for_uc)
#'
#' list_for_em_om <- list(
#' x1y0 = c(-2.18, 1.63, 0.23, 0.36),
#' x0y1 = c(-3.17, 0.22, 1.60, 0.40),
#' x1y1 = c(-4.76, 1.82, 1.83, 0.72)
#' )
#'
#' bp_em_om <- bias_params(coef_list = list_for_em_om)
#'
#' @export
bias_params <- function(coef_list) {
# Check if the input is a list (R's equivalent of a dictionary)
if (!is.list(coef_list)) {
stop("Input must be a list (dictionary).")
}
# Check if all keys are strings
if (!all(sapply(names(coef_list), is.character))) {
stop("All keys in the dictionary must be strings.")
}
# Check if all values are vectors
if (!all(sapply(coef_list, is.vector))) {
stop("All values in the dictionary must be vectors.")
}
# Check if key is an acceptable value
acceptable_values <- c(
"u", "x", "y", "s", "x1u0", "x0u1", "x1u1", "u1y0", "u0y1", "u1y1",
"x1y0", "x0y1", "x1y1"
)
for (name in names(coef_list)) {
if (!(name %in% acceptable_values)) {
stop(
paste0(
"Unacceptable value in coef_list. Acceptable values include: ",
"u, x, y, s, x1u0, x1u1, x1u1, u1y0, u0y1, u1y1, x1y0, x0y1, x1y1"
)
)
}
}
obj <- list(coef_list = coef_list)
class(obj) <- "bias_params"
return(obj)
}
#' @export
print.bias_params <- function(x, ...) {
cat("Bias Parameters\n")
cat("---------------------------------\n")
for (key in names(x$coef_list)) {
cat("Model", key, "\n")
cat(" Coefs: ", paste(x$coef_list[[key]], collapse = ", "), "\n")
}
invisible(x)
}
Try the multibias package in your browser
Any scripts or data that you put into this service are public.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.