{"created":"2021-03-01T07:09:47.605753+00:00","id":48970,"links":{},"metadata":{"_buckets":{"deposit":"65cf65b6-8f12-4e58-b024-4351e4a94179"},"_deposit":{"id":"48970","owners":[],"pid":{"revision_id":0,"type":"depid","value":"48970"},"status":"published"},"_oai":{"id":"oai:repository.dl.itc.u-tokyo.ac.jp:00048970","sets":["62:828:7685:7686","9:504:831:7687:7688"]},"item_4_alternative_title_1":{"attribute_name":"その他のタイトル","attribute_value_mlt":[{"subitem_alternative_title":"Investment with deep learning"}]},"item_4_biblio_info_7":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicIssueDates":{"bibliographicIssueDate":"2017-07-01","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"1","bibliographicPageEnd":"40","bibliographicPageStart":"2","bibliographicVolumeNumber":"82","bibliographic_titles":[{"bibliographic_title":"經濟學論集 = The Journal of Economics"}]}]},"item_4_description_5":{"attribute_name":"抄録","attribute_value_mlt":[{"subitem_description":"本稿では，ニューラルネットワークを用いた深層学習に基づく投資手法を考察する．特に，損失関数の特定化により様々な投資手法を構築できることに着目し，その例として，資産リターン（収益率）の時系列を入力し，教師付き深層学習（Supervised Deep Learning, SL）によるアノマリー検出を投資判断に活用する手法と，収益最大化に基づく投資判断を深層強化学習（Deep Reinforcement Learning, RL）により求める方法を示す．また，各資産のリターンに関する学習結果を，複数資産を対象とするポートフォリオ運用についても活用する．　さらに，現実の市場データを用いた検証を行い，対象資産が同じであっても，月次／日次リターン等の入力データの種類，学習手法の選択，投資期間中の再学習の有無，ネットワークの層数，中間層のユニット数等の外生的に設定する項目により投資パフォーマンスが大きく異なることを例示する．なかでも，RLに基づく投資判断が，ポートフォリオ運用において相対的には安定して良好なパフォーマンスを示すことが明らかとなる．","subitem_description_type":"Abstract"},{"subitem_description":"This paper considers investment methods with deep learning in neural networks. In particular, as one can create various investment strategies by different specifications of a loss function, the current work presents two examples based on return anomalies detected by supervised deep learning (SL) or profit maximization by deep reinforcement learning (RL). It also applies learning of individual asset return dynamics to portfolio strategies. Moreover, an empirical study shows that the investment performance are quite sensitive to exogenously specified items such as frequency of input data (e.g.monthly or daily returns), selection of a learning method, update of learning, number of layers in a network and number of units in intermediate layers. Especially, it is observed that RL provides relatively fine records in portfolio investment.","subitem_description_type":"Abstract"}]},"item_4_description_6":{"attribute_name":"内容記述","attribute_value_mlt":[{"subitem_description":"論文/Article","subitem_description_type":"Other"}]},"item_4_full_name_3":{"attribute_name":"著者別名","attribute_value_mlt":[{"nameIdentifiers":[{"nameIdentifier":"145511","nameIdentifierScheme":"WEKO"}],"names":[{"name":"FUKUI, Takaya"}]},{"nameIdentifiers":[{"nameIdentifier":"145512","nameIdentifierScheme":"WEKO"}],"names":[{"name":"TAKAHASHI, Akihiko"}]}]},"item_4_identifier_registration":{"attribute_name":"ID登録","attribute_value_mlt":[{"subitem_identifier_reg_text":"10.15083/00074095","subitem_identifier_reg_type":"JaLC"}]},"item_4_publisher_20":{"attribute_name":"出版者","attribute_value_mlt":[{"subitem_publisher":"東京大学大学院経済学研究科"}]},"item_4_select_14":{"attribute_name":"著者版フラグ","attribute_value_mlt":[{"subitem_select_item":"publisher"}]},"item_4_source_id_10":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN00070411","subitem_source_identifier_type":"NCID"}]},"item_4_source_id_8":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"0022-9768","subitem_source_identifier_type":"ISSN"}]},"item_4_text_21":{"attribute_name":"出版者別名","attribute_value_mlt":[{"subitem_text_value":"Graduate School of Economics, The University of Tokyo"}]},"item_4_text_4":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"みずほ証券株式会社"},{"subitem_text_value":"東京大学大学院経済学研究科"}]},"item_creator":{"attribute_name":"著者","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"福井, 貴也"}],"nameIdentifiers":[{"nameIdentifier":"145509","nameIdentifierScheme":"WEKO"}]},{"creatorNames":[{"creatorName":"高橋, 明彦"}],"nameIdentifiers":[{"nameIdentifier":"145510","nameIdentifierScheme":"WEKO"}]}]},"item_files":{"attribute_name":"ファイル情報","attribute_type":"file","attribute_value_mlt":[{"accessrole":"open_date","date":[{"dateType":"Available","dateValue":"2017-12-18"}],"displaytype":"detail","filename":"econ0820102.pdf","filesize":[{"value":"3.0 MB"}],"format":"application/pdf","licensetype":"license_note","mimetype":"application/pdf","url":{"label":"econ0820102.pdf","url":"https://repository.dl.itc.u-tokyo.ac.jp/record/48970/files/econ0820102.pdf"},"version_id":"c326c2df-aa1e-458e-86eb-28f9d5f0d8c5"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"深層学習","subitem_subject_scheme":"Other"},{"subitem_subject":"教師付き深層学習","subitem_subject_scheme":"Other"},{"subitem_subject":"深層強化学習","subitem_subject_scheme":"Other"},{"subitem_subject":"ニューラルネットワーク","subitem_subject_scheme":"Other"},{"subitem_subject":"投資手法","subitem_subject_scheme":"Other"},{"subitem_subject":"ポートフォリオ","subitem_subject_scheme":"Other"},{"subitem_subject":"deep learning","subitem_subject_scheme":"Other"},{"subitem_subject":"supervised deep learning","subitem_subject_scheme":"Other"},{"subitem_subject":"deep reinforcement learning","subitem_subject_scheme":"Other"},{"subitem_subject":"neural network","subitem_subject_scheme":"Other"},{"subitem_subject":"investment method","subitem_subject_scheme":"Other"},{"subitem_subject":"portfolio","subitem_subject_scheme":"Other"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourcetype":"departmental bulletin paper","resourceuri":"http://purl.org/coar/resource_type/c_6501"}]},"item_title":"深層学習を用いた投資手法","item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"深層学習を用いた投資手法"}]},"item_type_id":"4","owner":"1","path":["7686","7688"],"pubdate":{"attribute_name":"公開日","attribute_value":"2017-12-18"},"publish_date":"2017-12-18","publish_status":"0","recid":"48970","relation_version_is_last":true,"title":["深層学習を用いた投資手法"],"weko_creator_id":"1","weko_shared_id":2},"updated":"2022-12-19T04:23:36.916153+00:00"}