Initial CRAN release.
localAteDS() aggregate function computes the local ATE estimate and,
optionally, the variance statistics (arm-specific residual sums of squares
and within-server projected scatter) needed for the delta-method variance.
Set compute_variance = FALSE to limit disclosure to the three scalars
strictly needed for the point estimate (n_k, n_k_1, tau_k). The
heterogeneous argument (default TRUE) additionally returns covariate
column sums, allowing the client to assemble the global covariate mean
for the leverage-corrected variance estimator.
addStudyIdDS() assign function appends K-1 integer study-dummy columns
to a server-side data frame, supporting the adjusted generalised difference
estimator when study membership directly shifts the outcome.
All functions enforce DataSHIELD disclosure controls: privacy control
level is checked via dsBase::checkPermissivePrivacyControlLevel() and
minimum subset size is validated against nfilter.subset from
dsBase::listDisclosureSettingsDS().