{"id":"W3154200574","doi":"10.2139/ssrn.3812473","title":"Deep Reinforcement Learning for Algorithmic Trading","year":2021,"lang":"en","type":"article","venue":"SSRN Electronic Journal","topic":"Stock Market Forecasting Methods","field":"Decision Sciences","cited_by":14,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"","keywords":"Reinforcement learning; Foreign exchange market; Arbitrage; Currency; Computer science; Markov chain; Q-learning; Foreign exchange; Artificial intelligence; Exchange rate; Machine learning; Economics; Financial economics; Monetary economics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001375369,0.0006113788,0.001002059,0.0004421821,0.0002992852,0.0008909527,0.001022276,0.001331943,0.003896656],"category_scores_gemma":[0.006225079,0.0004371395,0.0003778409,0.000404744,0.0009081335,0.001096216,0.001068731,0.002121585,0.0003841271],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009126494,"about_ca_system_score_gemma":0.0009758534,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003815208,"about_ca_topic_score_gemma":0.003763098,"domain_scores_codex":[0.9996629,0.0001488482,0.00001882212,0.00005987046,0.00006546151,0.00004418699],"domain_scores_gemma":[0.9972718,0.002121859,0.00015643,0.0001326554,0.0002136153,0.0001035878],"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":[0.00006377132,0.00007868206,0.0007907627,0.00006979678,0.00004066988,0.00003783518,0.00003487588,0.8874043,0.0006175846,0.0462032,0.001554977,0.0631036],"study_design_scores_gemma":[0.000004217375,0.000007173996,0.00003223857,0.000003978505,0.000001998419,0.000002228099,0.000001185275,0.9876151,0.00005262018,0.01216559,0.0001121683,0.000001503574],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.028209,0.001010793,0.9660722,0.0006155436,0.000116828,0.00003232709,0.000051413,0.000310113,0.003581808],"genre_scores_gemma":[0.8801697,0.0006010841,0.1107058,0.0002290335,0.0001148443,0.0001401938,0.0001144165,0.00007695051,0.007847901],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003896656,"threshold_uncertainty_score":0.0130356,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06310182009114429,"score_gpt":0.3757627060276205,"score_spread":0.3126608859364762,"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."}}