psdata_bin_retrospective_sp: Retrospective Superpopulation Dataset for PSweight Analysis

In PSweight: Propensity Score Weighting for Causal Inference with Observational Studies and Randomized Trials

Retrospective Superpopulation Dataset for PSweight Analysis

Description

A simulated dataset representing the retrospective superpopulation (i.e. the full dataset) for propensity score weighting analysis under a retrospective design. In this design, the sampling probability is influenced by both covariates and treatment assignment.

Usage

data(psdata_bin_retrospective_sp)

Format

A data frame with 1500 rows and 10 variables:

Y

Outcome variable.

trt

Binary treatment indicator (coded as 1 and 2).

cov1, cov2, cov3, cov4, cov5, cov6

Pre-treatment covariates.

sampling_prob

Sampling probability computed as a function of covariates and treatment (e.g., with an additional term depending on trt).

survey_weight

Survey weight computed as the inverse of sampling_prob.

sample_indicator

Indicator variable (0/1) denoting whether an observation was selected into the sample.

Details

In the retrospective design, the sampling probability is determined by both covariates and treatment assignment. This dataset represents the full superpopulation in which the sampling mechanism is retrospective.

Examples

data(psdata_bin_retrospective_sp)
head(psdata_bin_retrospective_sp)

PSweight documentation built on April 3, 2025, 10:27 p.m.