Package: DPrivStats 0.1.0

DPrivStats: Differentially Private Classical Statistical Inference

Implements differentially private (DP) versions of common classical statistical procedures, including descriptive statistics (mean, variance, quantiles, histograms), hypothesis tests (t-test, chi-square, Kolmogorov-Smirnov, one-way ANOVA), and regression (closed-form DP linear regression and DP-SGD for generalized linear models). Provides Laplace and Gaussian mechanisms with analytic calibration, exponential mechanism for medians, privacy-aware confidence intervals that account for both sampling and privacy noise, and privacy budget accounting via basic, advanced, and Renyi differential privacy (RDP) composition. Designed for official statistics and privacy-preserving data analysis research.

Authors:Mukul Bijalwan [aut, cre]

DPrivStats_0.1.0.tar.gz
DPrivStats_0.1.0.zip(r-4.7-x86_64)DPrivStats_0.1.0.zip(r-4.7-arm64)DPrivStats_0.1.0.zip(r-4.6-x86_64)DPrivStats_0.1.0.zip(r-4.6-arm64)DPrivStats_0.1.0.zip(r-4.5-x86_64)
DPrivStats_0.1.0.tgz(r-4.6-x86_64)DPrivStats_0.1.0.tgz(r-4.6-arm64)DPrivStats_0.1.0.tgz(r-4.5-x86_64)DPrivStats_0.1.0.tgz(r-4.5-arm64)
DPrivStats_0.1.0.tar.gz(r-4.7-arm64)DPrivStats_0.1.0.tar.gz(r-4.7-x86_64)DPrivStats_0.1.0.tar.gz(r-4.6-arm64)DPrivStats_0.1.0.tar.gz(r-4.6-x86_64)
DPrivStats_0.1.0.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
DPrivStats/json (API)

# Install 'DPrivStats' in R:
install.packages('DPrivStats', repos = c('https://mukulbijalwan.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/mukulbijalwan/dprivstats/issues

Uses libs:
  • c++– GNU Standard C++ Library v3
Datasets:

On CRAN:

Conda:

cpp

4.65 score 6 scripts 383 downloads 41 exports 2 dependencies

Last updated from:f63ca49c7a. Checks:15 OK. Indexed: yes.

TargetResultTimeFilesSyslog
linux-devel-arm64OK93
linux-devel-x86_64OK117
source / vignettesOK180
linux-release-arm64OK89
linux-release-x86_64OK120
macos-release-arm64OK175
macos-release-x86_64OK361
macos-oldrel-arm64OK99
macos-oldrel-x86_64OK302
windows-devel-arm64OK300
windows-devel-x86_64OK143
windows-release-arm64OK338
windows-release-x86_64OK103
windows-oldrel-x86_64OK80
wasm-releaseOK87

Exports:advanced_composition_epsilonanalytic_gaussian_mechanismanalytic_gaussian_sigmabasic_compositioncan_spendclip_gradientscompare_compositioncompare_utilitycpp_clip_gradientscpp_rlaplacedp_anovadp_chisq_testdp_confintdp_glmdp_histogramdp_ks_testdp_lmdp_lm_sensitivitydp_meandp_mediandp_quantiledp_t_testdp_utility_diagnosticsdp_varianceempirical_delta_from_lossesepsilon_to_rdpexponential_mechanismgaussian_mechanismgaussian_sigmalaplace_mechanismlaplace_plr_quantilelaplace_plr_tailnew_privacy_budgetrdp_compositionrdp_to_epsilonrlaplacesimulate_datasimulate_gaussian_lossesspendstat_sensitivitiesvalidate_coverage

Dependencies:MASSRcpp

Getting Started with DPrivStats
Data | Privacy budget | DP descriptive statistics | DP t-test | Budget accounting
Rendered fromgetting-started.Rmdusingknitr::rmarkdown

Last update: 2026-08-22
Started: 2026-08-22

Official Statistics Workflow with DPrivStats
Combining releases under one budget | Post-processing constraints | Composition comparison | Disclosure risk intuition
Rendered fromofficial-statistics.Rmdusingknitr::rmarkdown

Last update: 2026-08-22
Started: 2026-08-22

DP Regression and Privacy-Aware Confidence Intervals
Closed-form DP linear regression | Privacy-aware confidence intervals | DP-SGD for GLMs | Coverage validation
Rendered fromregression-guide.Rmdusingknitr::rmarkdown

Last update: 2026-08-22
Started: 2026-08-22

Readme and manuals

Help Manual

Help pageTopics
Advanced compositionadvanced_composition_epsilon
Analytic Gaussian mechanismanalytic_gaussian_mechanism
Analytic Gaussian calibration (Balle & Wang, 2018)analytic_gaussian_sigma
Basic (sequential) compositionbasic_composition
Check whether a release fits in the remaining budgetcan_spend
Per-sample gradient clippingclip_gradients
Compare composition rules over a workflowcompare_composition
Compare utility of non-private and private fitscompare_utility
Confint interface for dp_lm objectsconfint.dp_lm
Fast per-sample gradient clipping (C++)cpp_clip_gradients
Fast Laplace sampling (C++)cpp_rlaplace
DP one-way ANOVA F-testdp_anova
DP chi-square test of independencedp_chisq_test
Privacy-aware confidence intervalsdp_confint
DP-SGD for generalized linear modelsdp_glm
DP histogramdp_histogram
DP Kolmogorov-Smirnov testdp_ks_test
Differentially private linear regressiondp_lm
Global L2 sensitivity of OLS coefficientsdp_lm_sensitivity
DP meandp_mean
DP median via the exponential mechanismdp_median
DP quantile via the exponential mechanismdp_quantile
DP two-sample t-testdp_t_test
Utility diagnostics for DP estimatorsdp_utility_diagnostics
DP variancedp_variance
Empirical delta from simulated privacy lossesempirical_delta_from_losses
Convert epsilon-DP to zCDP (RDP) parameterepsilon_to_rdp
Census-like example microdataexample_microdata
Exponential mechanism samplerexponential_mechanism
Gaussian mechanismgaussian_mechanism
Classic Gaussian noise scalegaussian_sigma
Laplace mechanismlaplace_mechanism
Privacy loss random variables for the Laplace mechanismlaplace_plr_quantile laplace_plr_tail
Create a privacy budget objectnew_privacy_budget
Print method for dp_confint objectsprint.dp_confint
Print method for dp_estimate objectsprint.dp_estimate
Print method for dp_glm objectsprint.dp_glm
Print method for dp_htest objectsprint.dp_htest
Print method for dp_lm objectsprint.dp_lm
Print method for privacy_budget objectsprint.privacy_budget
Print method for summary.dp_lm objectsprint.summary.dp_lm
RDP compositionrdp_composition
Convert zCDP (RDP) parameter to (epsilon, delta)-DPrdp_to_epsilon
Sample Laplace noiserlaplace
Simulate synthetic microdatasimulate_data
Simulate privacy losses of Gaussian releasessimulate_gaussian_losses
Spend privacy budgetspend
Sensitivity of common bounded-data statisticsstat_sensitivities
Summarize a dp_lm fitsummary.dp_lm
Validate confidence-interval coverage by Monte Carlovalidate_coverage