calibration.plot: Calibration plot

In CHMMaas/HTEPredictionMetrics: Heterogeneous treatment effect prediction metrics

Calibration plot

Description

This function makes the calibration plot of matched pairs, i.e. the agreement between observed and predicted pairwise treatment effect. Please note, this function is only applicable for binary outcomes.

Usage

calibration.plot(
  matched.patients = NULL,
  g = 5,
  plot.CI = FALSE,
  show = TRUE,
  ...
)

Arguments

matched.patients	dataframe; dataframe of matched patients, which contains predicted treatment effect of matched patients (matched.tau.hat) and observed treatment effect (matched.tau.obs) of matched patients
g	integer; indicating the number of groups in calibration plot
plot.CI	boolean; TRUE if you want to plot the confidence interval of the calibration plot of predicted versus observed treatment effect of matched patients
show	boolean; TRUE if you want to show the plot; FALSE if not
...	additional arguments for loess function from loess package

Value

The output of the E.for.benefit function is a "list" with the following components.

matched.patients

a dataframe containing the matched patients.

build.plot

ggplot2 object of the plot.

quantiles

of treatment effect.

se.quantiles

standard error of quantiles of treatment effect.

Examples

library(HTEPredictionMetrics)
set.seed(1)
n <- 100
Y <- sample(0:1, n, replace=TRUE)
W <- sample(0:1, n, replace=TRUE)
X <- matrix(rnorm(n*4), n, 4)
p.0 <- runif(n)
p.1 <- runif(n)
tau.hat <- runif(n)
g <- 5 # number of groups in calibration plot

# OPTION 1
matched.patients <- match.patients(Y=Y, W=W, X=X,
                                   p.0=p.0, p.1=p.1, tau.hat=tau.hat,
                                   CI=FALSE, message=TRUE,
                                   measure="nearest", distance="mahalanobis",
                                   estimand=NULL)
CP <- calibration.plot(matched.patients=matched.patients$df.matched.patients, g=g,
                 plot.CI=TRUE, show=TRUE)

# OPTION 2
EB.out <- E.for.Benefit(Y=Y, W=W, X=X, p.0=p.0, p.1=p.1, tau.hat=tau.hat,
                        CI=TRUE, nr.bootstraps=100, message=TRUE,
                        matched.patients=NULL,
                        measure="nearest", distance="mahalanobis",
                        estimand=NULL)
CP <- calibration.plot(matched.patients=EB.out$matched.patients, g=g,
                 plot.CI=TRUE, show=TRUE)

# EDIT THE CALIBRATION PLOT USING GGPLOT
CP$build.plot <- CP$build.plot+ggplot2::scale_x_continuous(limits=c(-1, 1))
show(CP)

CHMMaas/HTEPredictionMetrics documentation built on June 2, 2025, 5:04 a.m.