{"id":"W2889374799","doi":"10.1007/s10742-018-0187-2","title":"Causal inference for multi-level treatments with machine-learned propensity scores","year":2018,"lang":"en","type":"article","venue":"Health Services and Outcomes Research Methodology","topic":"Advanced Causal Inference Techniques","field":"Mathematics","cited_by":11,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo","funders":"Social Sciences and Humanities Research Council of Canada","keywords":"Propensity score matching; Causal inference; Inverse probability; Inverse probability weighting; Multinomial logistic regression; Statistics; Econometrics; Observational study; Logistic regression; Matching (statistics); Computer science; Estimator; Bayesian probability; Average treatment effect; Mathematics; Posterior probability","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":[0.0507329,0.00189765,0.005038728,0.004151046,0.00156406,0.004291061,0.005457031,0.004392186,0.01765967],"category_scores_gemma":[0.1805974,0.002539259,0.005261331,0.004471983,0.004050018,0.006628064,0.004500949,0.008481104,0.001409672],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003079888,"about_ca_system_score_gemma":0.005442534,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008350644,"about_ca_topic_score_gemma":0.008583022,"domain_scores_codex":[0.9677829,0.02523814,0.00123855,0.003459982,0.001436323,0.0008440358],"domain_scores_gemma":[0.737551,0.2428729,0.006503494,0.009794982,0.002123032,0.001154584],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.000664085,0.0005943732,0.008638517,0.001011539,0.001785428,0.0003467842,0.0004466784,0.2612616,0.0002469391,0.5985048,0.007246933,0.1192522],"study_design_scores_gemma":[0.0002269991,0.00005406824,0.0007958704,0.0001012871,0.0002261294,0.00005570704,0.00003107301,0.4518765,0.0001461227,0.545364,0.001094393,0.00002775722],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.005778077,0.0005850529,0.9902275,0.00160146,0.0001153867,0.0001975621,0.0006051234,0.0002903465,0.0005995351],"genre_scores_gemma":[0.430194,0.001917938,0.5516037,0.001624854,0.001169528,0.00317046,0.002721576,0.0002374693,0.007360514],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.0507329,"threshold_uncertainty_score":0.2683044,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.8430232517073458,"score_gpt":0.6674423948286655,"score_spread":0.1755808568786803,"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."}}