{"id":"W3143149325","doi":"10.6084/m9.figshare.14381684.v1","title":"Confounding adjustment methods for multi-level treatment comparisons under lack of positivity and unknown model specification","year":2021,"lang":"en","type":"article","venue":"Figshare","topic":"Advanced Causal Inference Techniques","field":"Mathematics","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"Centre hospitalier universitaire de Québec; Université Laval","funders":"Fonds de Recherche du Québec - Santé; Natural Sciences and Engineering Research Council of Canada","keywords":"Covariate; Estimator; Context (archaeology); Matching (statistics); Observational study; Propensity score matching; Confounding; Specification; Average treatment effect; Variance (accounting); Computer science; Statistics; Econometrics; Outcome (game theory); Contrast (vision); Mathematics; Machine learning; Artificial intelligence","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.09018072,0.001290449,0.002217255,0.002184871,0.001250591,0.001716987,0.003865137,0.002342857,0.008902999],"category_scores_gemma":[0.2969404,0.0008560784,0.00409952,0.00256277,0.002777094,0.003345116,0.004114085,0.003385524,0.0006467522],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001250594,"about_ca_system_score_gemma":0.003367845,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001949031,"about_ca_topic_score_gemma":0.002159166,"domain_scores_codex":[0.9213755,0.06632192,0.002641998,0.004639983,0.004269041,0.0007515383],"domain_scores_gemma":[0.7822056,0.1769954,0.01229705,0.02323755,0.004533934,0.0007305204],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.001790422,0.0003302663,0.0211693,0.001417352,0.004028374,0.0003883925,0.001103272,0.072432,0.002042157,0.3368576,0.005490904,0.55295],"study_design_scores_gemma":[0.001191717,0.00127779,0.010912,0.000412023,0.001161809,0.0003851706,0.0002433033,0.3615605,0.004057534,0.602942,0.01570702,0.0001491139],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.004473659,0.0002539723,0.9939148,0.0002679235,0.00007471644,0.0003179262,0.00008641632,0.000215977,0.0003946768],"genre_scores_gemma":[0.1778839,0.0002965188,0.8178732,0.0004423484,0.0001552879,0.00194411,0.0002856272,0.000158935,0.0009601173],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.09018072,"threshold_uncertainty_score":0.476927,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.8348849125377892,"score_gpt":0.5963865798411815,"score_spread":0.2384983326966077,"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."}}