{"id":"W4388558959","doi":"10.1002/9781119712206.ch20","title":"On Using Harmonized Data in Statistical Analysis: Notes of Caution","year":2023,"lang":"en","type":"other","venue":"","topic":"Forecasting Techniques and Applications","field":"Decision Sciences","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Weighting; Set (abstract data type); Computer science; Data science; Data set; Data mining; Missing data; Operations research; Machine learning; Artificial intelligence; Engineering","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001232565,0.00009805886,0.0003692729,0.001039423,0.00001780071,0.0000396287,0.0007654098,0.0001175595,0.002628874],"category_scores_gemma":[0.002655259,0.00006994457,0.00004808438,0.002087489,0.00007006175,0.00002750196,0.0002298679,0.00009510555,0.000115683],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000207682,"about_ca_system_score_gemma":0.00004064195,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005644796,"about_ca_topic_score_gemma":0.006994401,"domain_scores_codex":[0.9982796,0.0000638464,0.0004625475,0.0004830033,0.000604657,0.0001063636],"domain_scores_gemma":[0.9968366,0.001509491,0.0002712012,0.001312053,0.00004109809,0.00002956595],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002387283,0.0001913617,0.003270096,0.00001533229,0.0002280553,0.00001033402,0.00003367713,0.0007266542,0.00007996397,0.1081729,0.85295,0.0342977],"study_design_scores_gemma":[0.0003770143,0.00007151371,0.005969276,0.0002995635,0.0005256027,0.000001168623,0.00008106133,0.8373791,0.0001327686,0.0840846,0.07054483,0.0005335025],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0003546647,0.00002446912,0.9202664,0.0001597329,0.00006151306,0.0002935791,0.001918578,0.0003387266,0.07658234],"genre_scores_gemma":[0.1689632,0.00007083681,0.6079092,0.000101804,0.0001214609,0.00003922567,0.0008423395,0.0005449999,0.2214069],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.8366525,"threshold_uncertainty_score":0.9982828,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4979854027115637,"score_gpt":0.5258167326672998,"score_spread":0.02783132995573612,"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."}}