dpVariancePostProcess: Post-processing on differentially private variance, called...

In IQSS/PSI-Library: Differentially Private Statistical Releases for Privacy Preservation

Description

Post-processing on differentially private variance, called within the .self$mechanism$evaluate function, which in turn is called within dpVariance$release.

Arguments

out

Input dataframe that describes the differentially private release that was created in .self$mechanism$evaluate. This dataframe will have at least one pre-existing attribute, out$release, which is a numeric value of length one that is the differentially private variance. This function then post-processes the standard deviation from the variance and adds it to the dataframe as out$std, as long as the variance is positive. Additionally, known portions of the input such as the variable name and the accuracy of the DP variance and the epsilon value may be appended with no extra privacy loss.

Value

Dataframe out, updated to include post-processed values.

IQSS/PSI-Library documentation built on Feb. 15, 2020, 9:03 p.m.