{"id":"W4381190793","doi":"10.1111/jori.12436","title":"Improving risk classification and ratemaking using mixture‐of‐experts models with random effects","year":2023,"lang":"en","type":"article","venue":"Journal of Risk & Insurance","topic":"Bayesian Methods and Mixture Models","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Underwriting; Econometrics; Logistic regression; Profitability index; Actuarial science; Computer science; Categorization; Mixed logit; Economics; Mathematics; Statistics; Artificial intelligence; Finance","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":[],"consensus_categories":[],"category_scores_codex":[0.001781161,0.0001674317,0.0004387258,0.0002356865,0.0001741374,0.0001053517,0.0003502243,0.00008229051,2.050431e-7],"category_scores_gemma":[0.0003573235,0.0001160305,0.0001005096,0.0005868922,0.00006077208,0.0009178109,0.00006450724,0.0003104337,2.999733e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003858454,"about_ca_system_score_gemma":0.0000894666,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000500963,"about_ca_topic_score_gemma":0.000004682966,"domain_scores_codex":[0.9982244,0.0004558135,0.0004831833,0.0002465151,0.0003622391,0.0002278787],"domain_scores_gemma":[0.9972982,0.0006015989,0.001313778,0.0003014889,0.000385654,0.00009929143],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0007759784,0.0001118223,0.02679104,0.0003935318,0.0002352985,0.0001662973,0.006730735,0.0265559,0.07582957,0.01112863,0.0001213357,0.8511599],"study_design_scores_gemma":[0.003012261,0.000205073,0.03028377,0.0005985901,0.00007646039,0.0001563413,0.00005300689,0.9353194,0.007545422,0.02245201,0.00003966332,0.0002579739],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4197004,0.001241081,0.5786652,0.00002251416,0.0002423825,0.00009052615,0.000002163683,0.00001877089,0.00001690091],"genre_scores_gemma":[0.6690376,0.0007087944,0.3301302,0.00001657015,0.00009095472,0.000001881248,1.095857e-7,0.00001106103,0.000002789656],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9087635,"threshold_uncertainty_score":0.4731586,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0192545522689254,"score_gpt":0.2637129551929957,"score_spread":0.2444584029240703,"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."}}