{"id":"W4387952061","doi":"10.1016/j.aap.2023.107239","title":"Even correctly specified and well-estimated regression models can mislead","year":2023,"lang":"en","type":"article","venue":"Accident Analysis & Prevention","topic":"Bayesian Modeling and Causal Inference","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"","keywords":"Variable (mathematics); Regression analysis; Regression; Variables; Interpretation (philosophy); TRACE (psycholinguistics); Econometrics; Constant (computer programming); Regression toward the mean; Causality (physics); Linear regression; Function (biology); Statistics; Computer science; Mathematics","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005743047,0.0001833666,0.0002987248,0.0005432431,0.0001913629,0.0002476515,0.0004972931,0.00009491289,0.00004968073],"category_scores_gemma":[0.00002870568,0.0001657832,0.0001916061,0.002258883,0.00002257576,0.0005737739,0.0002573878,0.000137062,0.00009869166],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004393794,"about_ca_system_score_gemma":0.00004096547,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002360974,"about_ca_topic_score_gemma":0.0008022898,"domain_scores_codex":[0.9982116,0.0001554927,0.0003706662,0.0005786062,0.000385521,0.0002980913],"domain_scores_gemma":[0.9989644,0.00005873069,0.0001876684,0.0005533881,0.000113254,0.0001225418],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001205105,0.0004299357,0.1679761,0.00003258874,0.002764412,0.0001730282,0.005853417,0.1976301,0.006118791,0.03797863,0.01817685,0.5627457],"study_design_scores_gemma":[0.0001874339,0.00003485167,0.03486064,0.00007761065,0.0002821466,0.000003935219,0.00004956547,0.9333342,0.000669494,0.03025863,0.00003210982,0.0002093391],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3531999,0.0001832681,0.6448951,0.0005988296,0.0001411536,0.0001153153,2.004163e-7,0.0003526206,0.0005135917],"genre_scores_gemma":[0.9926302,0.0003476885,0.004156699,0.00005339295,0.00003416976,0.00001788314,0.00007475569,0.00001008819,0.002675169],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7357042,"threshold_uncertainty_score":0.6760442,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05590736184706821,"score_gpt":0.3167823341346346,"score_spread":0.2608749722875664,"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."}}