{"id":"W3209485542","doi":"10.1080/00031305.2021.2000495","title":"Comparative Probability Metrics: Using Posterior Probabilities to Account for Practical Equivalence in A/B tests","year":2021,"lang":"en","type":"article","venue":"The American Statistician","topic":"Statistical Methods and Bayesian Inference","field":"Mathematics","cited_by":12,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Mathematics; Statistics; Equivalence (formal languages); Posterior probability; Bayesian probability; Discrete mathematics","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.001186372,0.0002439,0.000767624,0.00009723903,0.0001515002,0.0001242303,0.0002495657,0.00003117548,0.00007166648],"category_scores_gemma":[0.02264453,0.0001829341,0.0000627429,0.001079166,0.0005891947,0.0001026401,0.0001715645,0.0002253553,0.00001027544],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002655622,"about_ca_system_score_gemma":0.0005647173,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003181767,"about_ca_topic_score_gemma":0.0006315158,"domain_scores_codex":[0.9972322,0.0007306961,0.0006154346,0.0005067688,0.0003673367,0.0005475927],"domain_scores_gemma":[0.9818524,0.01666597,0.0003039644,0.0005302334,0.0004994,0.0001479995],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0004156463,0.0005682789,0.002488759,0.0004036205,0.00006777311,0.00005992771,0.004614784,0.00002049251,0.002788653,0.9409472,0.0006368591,0.04698803],"study_design_scores_gemma":[0.0002435843,0.0004625357,0.01607747,0.0001244516,0.00008837808,0.00003956751,0.003470222,0.004710563,0.0008051866,0.9735093,0.0001432154,0.0003255125],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.205436,0.00001562577,0.79144,0.001179018,0.00008387363,0.001086911,0.0004647404,0.00003252271,0.0002613399],"genre_scores_gemma":[0.2337232,0.000001639912,0.7655104,0.0005017131,0.00005012004,0.0001693046,0.000003674608,0.00001923099,0.00002074825],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.04666252,"threshold_uncertainty_score":0.9855881,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2844714646661516,"score_gpt":0.5027176178365009,"score_spread":0.2182461531703493,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}