{"id":"W3196495100","doi":"10.1093/ije/dyab168.399","title":"286Collider-stratification bias when estimating variable importance using Random Forests","year":2021,"lang":"en","type":"article","venue":"International Journal of Epidemiology","topic":"Advanced Causal Inference Techniques","field":"Mathematics","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université du Québec à Montréal; McGill University","funders":"","keywords":"Causal inference; Outcome (game theory); Feature selection; Random forest; Inference; Collider; Variable (mathematics); Econometrics; Variables; Statistics; Computer science; Selection bias; Machine learning; Artificial intelligence; Mathematics; Physics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.1152031,0.001172077,0.001821985,0.001667147,0.001211291,0.003396874,0.002912083,0.00188002,0.003269714],"category_scores_gemma":[0.3136517,0.0008998364,0.00218709,0.001842103,0.003866813,0.002864079,0.002365745,0.003818769,0.0005490318],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001554213,"about_ca_system_score_gemma":0.002656812,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006277771,"about_ca_topic_score_gemma":0.005499809,"domain_scores_codex":[0.9280235,0.06084362,0.002157008,0.005101403,0.002901207,0.0009732629],"domain_scores_gemma":[0.5827863,0.3800875,0.01365447,0.01685625,0.005778552,0.0008369826],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001098789,0.0002957394,0.1929233,0.001795652,0.004034685,0.001129814,0.001711272,0.4218808,0.002190955,0.1835029,0.008444619,0.1809916],"study_design_scores_gemma":[0.0001679713,0.0001887369,0.01443467,0.0006238144,0.0004257176,0.0003660721,0.0002205992,0.6666353,0.002626257,0.3098376,0.004371608,0.0001016522],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07304034,0.001465536,0.9196826,0.001992438,0.0003040414,0.0003525435,0.0005014151,0.0006026994,0.002058338],"genre_scores_gemma":[0.8125312,0.0004244977,0.1839677,0.0008691146,0.0002778117,0.000528043,0.000512729,0.0001669379,0.0007219909],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8847969,"threshold_uncertainty_score":0.6092597,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4015642107846774,"score_gpt":0.4984844126900138,"score_spread":0.0969202019053364,"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."}}