Man pages for banditsCI
Bandit-Based Experiments and Policy Evaluation
| aw_estimate | Estimate policy value via non-contextual adaptive weighting. |
| aw_scores | Compute AIPW/doubly robust scores. |
| aw_var | Variance of policy value estimator via non-contextual... |
| calculate_balwts | Calculate balancing weight scores. |
| calculate_continuous_X_statistics | Estimate/variance of policy evaluation via contextual... |
| dot-check_A | Check Number of Observations for Inference |
| dot-check_first_batch | Check First Batch Validity |
| dot-check_shape | Check Shape Compatibility of Probability Objects |
| draw_thompson | Thompson Sampling draws. |
| estimate | Estimate/variance of policy evaluation via non-contextual... |
| generate_bandit_data | Generate classification data. |
| ifelse_clip | Clip lamb values between a minimum x and maximum y. |
| impose_floor | Impose probability floor. |
| LinTSModel | Linear Thompson Sampling model. |
| output_estimates | Policy evaluation with adaptively generated data. |
| plot_cumulative_assignment | Plot cumulative assignment for bandit experiment. |
| ridge_init | Ridge Regression Initialization for Arm Expected Rewards |
| ridge_muhat_lfo_pai | Leave-future-out ridge-based estimates for arm expected... |
| ridge_update | Updates ridge regression matrices. |
| run_experiment | Run an experiment using Thompson Sampling. |
| simple_tree_data | Generate simple tree data. |
| stick_breaking | Stick breaking function. |
| twopoint_stable_var_ratio | Calculate allocation ratio for a two-point stable-variance... |
| update_thompson | Update linear Thompson Sampling model. |
