Skip to contents

Doubly robust estimator of the average treatment effect for continuous data

Source: R/ATE_continuous.R
drmean.Rd

Doubly robust estimator of the average treatment effect between two treatments, which is the mean difference of treatment 1 over treatment 0 for continuous outcomes.

Usage

drmean(
  y,
  trt,
  x.cate,
  x.ps,
  ps.method = "glm",
  minPS = 0.01,
  maxPS = 0.99,
  interactions = TRUE
)

Arguments

y

A numeric vector of size n with each element representing the observed continuous outcome for each subject.

trt

A numeric vector (in {0, 1}) of size n with each element representing the treatment received for each subject.

x.cate

A numeric matrix of dimension n by p.cate with each column representing each baseline covariate specified in the outcome model for all subjects.

x.ps

A numeric matrix of dimension n by p.ps + 1 with a leading column of 1 as the intercept and each remaining column representing each baseline covariate specified in the propensity score model for all subjects

ps.method

A character value for the method to estimate the propensity score. Allowed values include one of: 'glm' for logistic regression with main effects only (default), or 'lasso' for a logistic regression with main effects and LASSO penalization on two-way interactions (added to the model if interactions are not specified in ps.model). Relevant only when ps.model has more than one variable.

minPS

A numerical value between 0 and 1 below which estimated propensity scores should be truncated. Default is 0.01.

maxPS

A numerical value between 0 and 1 above which estimated propensity scores should be truncated. Must be strictly greater than minPS. Default is 0.99.

interactions

A logical value indicating whether the outcome model should assume interactions between x and trt. If TRUE, interactions will be assumed only if at least 10 patients received each treatment option. Default is TRUE.

Value

Return a list of 4 elements:

  • mean.diff: A numeric value of the estimated mean difference.

  • mean.diff0: A numeric value of the estimated mean difference in treatment group 0.

  • mean.diff1: A numeric value of the estimated mean difference in treatment group 1.