{"id":"W2509712226","doi":"10.1016/j.jclinepi.2016.05.017","title":"Propensity score model overfitting led to inflated variance of estimated odds ratios","year":2016,"lang":"en","type":"article","venue":"Journal of Clinical Epidemiology","topic":"Advanced Causal Inference Techniques","field":"Mathematics","cited_by":60,"is_retracted":false,"has_abstract":false,"ca_institutions":"McGill University; Montreal Children's Hospital; Jewish General Hospital","funders":"Canadian Institutes of Health Research","keywords":"Overfitting; Propensity score matching; Statistics; Type I and type II errors; Confounding; Logistic regression; Inference; Econometrics; Mathematics; Computer science; Artificial intelligence; Artificial neural network","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":[],"category_scores_codex":[0.08315752,0.001267809,0.002298191,0.001774453,0.001016987,0.003741348,0.002386489,0.00231342,0.003445813],"category_scores_gemma":[0.3488279,0.001466667,0.002297476,0.002289312,0.002030086,0.002937096,0.002207991,0.007915907,0.0008134899],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001430674,"about_ca_system_score_gemma":0.002336859,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003937202,"about_ca_topic_score_gemma":0.003278274,"domain_scores_codex":[0.9587513,0.02743009,0.002396711,0.006323946,0.004147192,0.0009507301],"domain_scores_gemma":[0.7687367,0.1891984,0.005696596,0.02733321,0.008487498,0.0005476955],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0008959538,0.0003737923,0.05595977,0.0006720183,0.002201431,0.002829065,0.002909847,0.3070181,0.006458688,0.3427593,0.01592217,0.2619999],"study_design_scores_gemma":[0.0001353477,0.0001039614,0.008523117,0.0001333867,0.0004794504,0.001086408,0.00009027795,0.6747529,0.004419568,0.3055602,0.004611536,0.0001038653],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01655746,0.0002230571,0.9807373,0.0009974424,0.0001793021,0.00007112187,0.0001163699,0.0005405808,0.0005773167],"genre_scores_gemma":[0.6475312,0.0003010991,0.3464497,0.001610933,0.0003004792,0.0003429398,0.0004438855,0.0008112783,0.002208466],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9168425,"threshold_uncertainty_score":0.4397843,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.7352681527672233,"score_gpt":0.5929485520547376,"score_spread":0.1423196007124857,"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."}}