Experimental Evaluation of Algorithm-Assisted Human Decision-Making
Llama3 Recommendations (internal)
Gibbs sampler for the main analysis
tools:::Rd_package_title("aihuman")
Summary of APCE
Summary of APCE for frequentist analysis
Bootstrap for estimating variance of APCE
Bootstrap for estimating variance of APCE with random effects
Bootstrap for estimating variance of APCE with random effects
Calculate APCE
Compute APCE using frequentist analysis
Compute APCE using frequentist analysis with random effects
Calculate APCE using parallel computing
Calculate the delta given the principal stratum
Calculate diff-in-means estimates
Calculate diff-in-means estimates
Calculate the principal fairness
Calculate optimal decision & utility
Calculate the proportion of principal strata (R)
Compute Risk (AI v. Human)
Fit outcome/decision and propensity score models conditioning on the A...
Fit outcome/decision and propensity score models
Agreement of Human and AI Decision Makers
Compute Risk (Human+AI v. Human)
Compute Risk (Human+AI v. Human) for a Subgroup Defined by AI Recommen...
Compute Risk (Human+AI v. Human)
Crossfitting for nuisance functions
Pulling ggplot legend
NCA follow policy (internal; decreasing monotonicity)
NCA follow policy (internal; increasing monotonicity)
NCA provide policy (internal; decreasing monotonicity)
NCA provide policy (internal; increasing monotonicity)
Nuisance functions conditioning on AI (internal)
Nuisance functions (internal)
Visualize Agreement
Visualize Difference in Risk (AI v. Human)
Visualize Difference in Risk (Human+AI v. Human)
Visualize Difference in Risk (Human+AI v. Human)
Visualize Difference in Risk (Human+AI v. Human) for a Subgroup Define...
Visualize Preference
Plot APCE
Plot diff-in-means estimates
Plot diff-in-means estimates
Plot the principal fairness
Plot optimal decision
Plot the proportion of principal strata (R)
Plot conditional randomization test
Plot power analysis of conditional randomization test
Stacked barplot for the distribution of the decision given psa
Stacked barplot for the distribution of the decision given DMF recomme...
Plot utility difference
Plot utility difference with 95% confidence interval
Conduct conditional randomization test
Conduct power analysis of conditional randomization test
Table of Agreement
Test monotonicity
Test monotonicity with random effects
Visualize Agreement (internal)
Visualize Risk (AI v. Human; internal)
Visualize Risk (Human+AI v. Human; internal)
Visualize Risk (Human+AI v. Human; internal)
Visualize Preference (internal)
Provides statistical methods for analyzing experimental evaluation of the causal impacts of algorithmic recommendations on human decisions developed by Imai, Jiang, Greiner, Halen, and Shin (2023) <doi:10.1093/jrsssa/qnad010> and Ben-Michael, Greiner, Huang, Imai, Jiang, and Shin (2024) <doi:10.48550/arXiv.2403.12108>. The data used for this paper, and made available here, are interim, based on only half of the observations in the study and (for those observations) only half of the study follow-up period. We use them only to illustrate methods, not to draw substantive conclusions.