{"id":"W4311681269","doi":"10.22215/etd/2022-15153","title":"Deep Reinforcement Learning for Quantitative Finance: Time Series Forecasting using Proximal Policy Optimization","year":2022,"lang":"en","type":"dissertation","venue":"","topic":"Stock Market Forecasting Methods","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Carleton University","funders":"","keywords":"Software portability; Reinforcement learning; Computer science; Artificial intelligence; Machine learning; Sequence (biology); Deep learning; Space (punctuation); Feature (linguistics); Series (stratigraphy)","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":["metaresearch","metaepi_narrow","sts","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.00660091,0.0005582518,0.0008951629,0.001428892,0.001607919,0.0005327275,0.0008667018,0.0002902799,0.00386627],"category_scores_gemma":[0.04703107,0.0005122431,0.0003903367,0.002254655,0.000071143,0.0008178878,0.0002432843,0.0005612301,0.00001987539],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004235466,"about_ca_system_score_gemma":0.0009645556,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001215528,"about_ca_topic_score_gemma":0.00005006051,"domain_scores_codex":[0.9937466,0.0007508798,0.001672985,0.001159435,0.001954872,0.0007152238],"domain_scores_gemma":[0.9912641,0.004503098,0.002237024,0.0004992011,0.001401717,0.00009483549],"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":[0.00088131,0.00001755551,0.00004968403,0.00006559405,0.00004854919,0.00000292725,0.002757653,0.9649506,0.0001086253,0.002698494,0.0004181681,0.02800091],"study_design_scores_gemma":[0.0004174926,0.0006187365,0.00004306473,0.0001006477,0.00006885106,0.00001845602,0.007429961,0.984358,0.0003094601,0.003646699,0.002428349,0.0005603033],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.005836581,0.0001379075,0.942394,0.00007386126,0.001095488,0.001974937,0.00001020018,0.0001596427,0.0483174],"genre_scores_gemma":[0.005741111,0.00001465496,0.8050119,0.00005223611,0.0003044421,0.0004932606,0.001126523,0.0001555094,0.1871004],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.1387829,"threshold_uncertainty_score":0.9997329,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1328827027321908,"score_gpt":0.4344314375910162,"score_spread":0.3015487348588254,"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."}}