feat: add split linear regression

NAMESPACE

@@ -7,6 +7,7 @@ export(ds.boundedSumDP)
 export(ds.boundedVarianceDP)
 export(ds.chisqDP)
 export(ds.countDP)
+export(ds.linearRegressionDP)
 export(ds.maxDP)
 export(ds.medianDP)
 export(ds.minDP)

R/ds.linearRegressionDP.R

@@ -0,0 +1,31 @@
+source("R/utils.R")
+
+#' @title Differentially private count
+#'
+#' @param target_column target response
+#' @param input_columns explanatory variables
+#' @param epsilon privacy budget
+#' @param type a character string that represents the type of analysis to carry out.
+#' This can be set as \code{'combine'}, \code{'split'} or \code{'both'}.
+#' @param datasources a list of \code{\link{DSConnection-class}} 
+#' objects obtained after login. If the \code{datasources} argument is not specified
+#' the default set of connections will be used: see \code{\link{datashield.connections_default}}.
+#'
+#' @return \code{ds.countDP} returns to the client-side a list including: \cr
+#' \code{Count.by.Study} (private count) for each study (if \code{type = split} or \code{type = both}). \cr
+#' \code{Global.Count} (private count) and across all studies combined (if \code{type = combine} or \code{type = both}).
+#' @export
+
+ds.linearRegressionDP <- function(target_column, input_columns, epsilon, type="split", datasources=NULL) {
+
+  if (is.null(datasources)) {
+    datasources <- DSI::datashield.connections_find()
+  }
+
+  coefficients.split <- callAggregationMethod(datasources, paste0("linearRegressionDP(", target_column, ", ", input_columns, ", ", epsilon, ")"))
+  coefficients.combine <- NULL
+
+  if (type=="combine") return(list(Global.Coefficients=coefficients.combine))
+  if (type=="split") return(list(Coefficients.by.Study=coefficients.split))
+  if (type=="both") return(list(Coefficients.by.Study=coefficients.split,Global.Coefficients=coefficients.combine))
+}