{"id":"W4416897195","doi":"10.21203/rs.3.rs-8233569/v1","title":"Explainable deep learning for American rate and volatility models","year":2025,"lang":"","type":"preprint","venue":"Research Square","topic":"Stock Market Forecasting Methods","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Calgary","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Deep learning; Volatility (finance); Valuation (finance); Artificial neural network; Modular design; Range (aeronautics); Volatility smile; Valuation of options","routes":{"ca_aff":true,"ca_fund":true,"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","scholarly_communication","open_science","research_integrity"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.1268364,0.0009332021,0.002199118,0.002886948,0.003774473,0.002547429,0.002869761,0.0006826759,0.0004944946],"category_scores_gemma":[0.1748131,0.0008672015,0.0006998157,0.005298894,0.002629743,0.0007225286,0.009106928,0.005305531,0.00002951112],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008535015,"about_ca_system_score_gemma":0.002324403,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001606366,"about_ca_topic_score_gemma":0.0002374662,"domain_scores_codex":[0.9621421,0.02285277,0.002254596,0.004677983,0.004935557,0.003137045],"domain_scores_gemma":[0.8904017,0.0966896,0.0009696896,0.002856822,0.008075829,0.001006321],"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":[0.002123135,0.0001612472,0.01412441,0.001929342,0.0001439064,0.00001814868,0.004183515,0.02599313,0.0000506678,0.001674417,0.001080116,0.948518],"study_design_scores_gemma":[0.0006550648,0.0008051171,0.008192176,0.0009385122,0.00003738319,0.000003911664,0.009232624,0.7777452,0.0001028602,0.1814511,0.02025845,0.0005775485],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1467553,0.003143814,0.8143607,0.001371554,0.0008971957,0.00759649,0.0002253606,0.0001750991,0.02547452],"genre_scores_gemma":[0.8659623,0.001631799,0.08997026,0.00005197414,0.0003672484,0.002291429,0.00004766795,0.0001021894,0.03957508],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9479404,"threshold_uncertainty_score":0.9993778,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2716726328978632,"score_gpt":0.5247673669305998,"score_spread":0.2530947340327366,"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."}}