{"id":"W2987365550","doi":"10.1161/circoutcomes.119.005927","title":"Effect of Variable Selection Strategy on the Performance of Prognostic Models When Using Multiple Imputation","year":2019,"lang":"en","type":"article","venue":"Circulation Cardiovascular Quality and Outcomes","topic":"Statistical Methods and Bayesian Inference","field":"Mathematics","cited_by":32,"is_retracted":false,"has_abstract":true,"ca_institutions":"University Health Network; University of Toronto; Sunnybrook Hospital; TD Bank Group; Institute of Health Services and Policy Research; Toronto Rehabilitation Institute; Sunnybrook Health Science Centre","funders":"Medical Research Council; Canadian Institutes of Health Research","keywords":"Missing data; Imputation (statistics); Statistics; Logistic regression; Feature selection; Sample size determination; Sample (material); Variables; Regression analysis; Regression; Computer science; Data mining; Mathematics; 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":["metaresearch"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.2259339,0.00181589,0.001898867,0.001675113,0.001151027,0.00181506,0.001936109,0.001866971,0.00137089],"category_scores_gemma":[0.3402371,0.0005887884,0.003373926,0.003029092,0.001522414,0.00194758,0.001809929,0.002519554,0.0004830636],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008351452,"about_ca_system_score_gemma":0.00192713,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001970819,"about_ca_topic_score_gemma":0.001502825,"domain_scores_codex":[0.7904421,0.1919286,0.006779746,0.004366068,0.005556195,0.0009273696],"domain_scores_gemma":[0.4277693,0.5302129,0.01492999,0.01728598,0.00873726,0.001064527],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.008941718,0.0007526306,0.5814192,0.0009535062,0.01067417,0.001081109,0.001624753,0.1516107,0.002880727,0.003859523,0.005335503,0.2308665],"study_design_scores_gemma":[0.001858252,0.00607548,0.1410334,0.0009577415,0.004358613,0.001717608,0.0004822801,0.8137043,0.01040864,0.01603508,0.003128391,0.000240198],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.611006,0.004272738,0.3755807,0.003599734,0.0004406772,0.0006990451,0.0008918527,0.0009183335,0.00259091],"genre_scores_gemma":[0.9154653,0.000471632,0.0817106,0.0005330527,0.00009619498,0.0004070789,0.0008526521,0.0001772592,0.0002862334],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7740662,"threshold_uncertainty_score":0.9545614,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0995815895126265,"score_gpt":0.360958675853795,"score_spread":0.2613770863411685,"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."}}