{"id":"W4385825735","doi":"10.48550/arxiv.2308.06260","title":"ChatGPT-based Investment Portfolio Selection","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Stock Market Forecasting Methods","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Mitacs","keywords":"Portfolio; Portfolio optimization; Selection (genetic algorithm); Black–Litterman model; Investment strategy; Stock market; Stock (firearms); Computer science; Economics; Econometrics; Artificial intelligence; Replicating portfolio; Financial economics; Microeconomics; Engineering; Context (archaeology)","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.005416377,0.0004710786,0.0006525653,0.0016325,0.0002921147,0.0002582616,0.001915091,0.0005201952,0.0007493467],"category_scores_gemma":[0.004277734,0.0004818098,0.0005113563,0.003246136,0.0002065705,0.0001969293,0.001487368,0.0008598646,0.001042915],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005329199,"about_ca_system_score_gemma":0.0006806972,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003460093,"about_ca_topic_score_gemma":0.0001714071,"domain_scores_codex":[0.9949194,0.001108791,0.000634697,0.00212131,0.0006511498,0.0005647092],"domain_scores_gemma":[0.9942206,0.002277759,0.0008649087,0.001683116,0.0006156698,0.0003379845],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003028695,0.0002016194,0.09189659,0.00006788285,0.0002117138,0.0005651286,0.0001909673,0.8307646,0.00007146835,0.02894889,0.04073885,0.006039413],"study_design_scores_gemma":[0.000651767,0.0001157952,0.03393871,0.0001198414,0.0001490583,0.000006296124,0.0001589957,0.5869063,0.0002690244,0.3669227,0.01000853,0.0007529478],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5777322,0.00003149376,0.3889551,0.0002742264,0.003790705,0.0008526017,0.00005864584,0.0009513896,0.02735357],"genre_scores_gemma":[0.9589093,0.00001852612,0.0047841,0.0003459999,0.0002194808,0.000004937653,0.00002918102,0.00006733619,0.03562108],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3841711,"threshold_uncertainty_score":0.9997634,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3366996575910667,"score_gpt":0.3117637748696222,"score_spread":0.02493588272144448,"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."}}