Man pages for jongbinjung/undi
Test for Unjustified Disparate Impact

compute_bmCompute benchmark test for disparate impact of a policy
compute_otCompute outcome tests for disparate impact of a policy
compute_radCompute risk-adjusted disparate impact estimate for a policy
dplyr.policyGenerate a new policy after applying various dplyr verbs
estimate_policy(Re)Estimate treatment and response surface models for a...
fit_glmnetWrapper for fitting 'glmnet' models from a formula
fit_sgdWrapper for fitting 'sgd' models with standardized design...
gridsensFind parameter values that min/maximize sensitivity results...
modelsList of paired fit/pred methods by type (glm/glmnet/gbm)
optimsensFind parameter values that min/maximize sensitivity results...
overwrite_listOverwrites a list with named arguments
plot.policyPlot risk-treatment and calibration plots for a 'policy'
plot.sensPlot min/max sensitivity results for a policy from a 'sens'...
plot.sensitive_policyPlot changes in risk-treatment for a 'sensitive_policy'
policyCreate policy object
predict4mmWrapper to compute predictions from a model matrix
print.policy'print' method of object class 'policy'
print.sens'print' method of object class 'sens' (parent of 'optimsens',...
rad_controlGiven a policy and (optional) controls, generate a...
sensitivitySensitivity to unobserved confounders with specified...
sensitize.policyCompute the sensitivity-adjusted estimates of predicted...
summary.sensSummary of sens object
synthesizeGenerate synthetic sample of a policy
undiundi: Testing for unjustified disparate impact
jongbinjung/undi documentation built on May 8, 2019, 11:56 p.m.