Description Usage Arguments Details Value
View source: R/utilities-histogram.R
Produce differentially private Herfindahl index for categorical types of data.
1 | histogramPostHerfindahl(release, n)
|
release |
A numeric vector with a noisy estimate of bin counts. |
n |
A numeric vector of length one specifying the number of observations in in the data. |
Let s_i be the percentage of data points in category i in the result histogram. Then, the Herfindahl index is defined to be the sum of the squares of s_i over all categories. Since the percentage of data points in category i may be computed directly from the bin counts in a histogram, a differentially private Herfindahl index may be calculated from a noisy histogram with no additional privacy loss incurred.
Herfindahl index.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.