Difference in difference + AIPW + external control borrowing
Usage
DID_EC_AIPW(
data,
outcome_col_name,
trial_status_col_name,
treatment_col_name,
covariates_col_name,
T_cross,
model_form_piS = "",
model_form_piA = "",
model_form_mu0_ext = "",
Bootstrap = FALSE,
R = 500,
bootstrap_CI_type = "bca",
alpha = 0.05,
quiet = TRUE
)Arguments
- data
data.frame. The input data containing the outcome, trial status, treatment, and covariates.
- outcome_col_name
character. The column name for the outcome variable in the data.
- trial_status_col_name
character. The column name for the trial status variable in the data (indicating RCT vs external control).
- treatment_col_name
character. The column name for the treatment variable in the data.
- covariates_col_name
character vector. The column names for the covariates in the data.
- T_cross
numeric. The time point that separates the placebo-control period and the follow-up period.
- model_form_piS
character. The model formula for the selection model (S).
- model_form_piA
character. The model formula for the treatment model (A).
- model_form_mu0_ext
character. The model formula for the outcome model in the external data (mu0_ext).
- Bootstrap
logical. Whether to use bootstrap for inference.
- R
numeric. The number of bootstrap replications.
- bootstrap_CI_type
character. The type of bootstrap confidence interval to compute (e.g., "bca", "norm", "perc", "basic", "stud").
- alpha
numeric. The significance level for confidence intervals.
- quiet
Logical. If
TRUE, suppress printed output.