博碩士論文 90423032 完整後設資料紀錄

DC 欄位 語言
DC.contributor資訊管理學系zh_TW
DC.creator黃春雅zh_TW
DC.creatorCheun-Yea Hwangen_US
dc.date.accessioned2003-6-26T07:39:07Z
dc.date.available2003-6-26T07:39:07Z
dc.date.issued2003
dc.identifier.urihttp://ir.lib.ncu.edu.tw:444/thesis/view_etd.asp?URN=90423032
dc.contributor.department資訊管理學系zh_TW
DC.description國立中央大學zh_TW
DC.descriptionNational Central Universityen_US
dc.description.abstract本研究將投資行為視為投資人與市場的賽局,考量投資人參與此賽局的經驗,應用遺傳演算法發展投資策略。賽局理論為一種分析策略性行為的方法,而所謂策略性行為是將對方會如何反應加入考量,而最終制定決策的行為。本研究探討三種不同型式的投資策略,當日沖銷、 交易策略、 資金管理。依此種賽局架構發展出的投資策略績效如下:(1) 當日沖銷策略,訓練期GA策略績效僅劣於全勝策略,而測試期的GA策略績效居中,但亦遠優於全敗策略;(2) 交易策略,GA策略在訓練期的績效,勝過原本的交易策略,但在測試期則不明顯優於原交易策略;(3) 資金管理,訓練期間 GA 所發展之資金管理策略有較高的期末淨值,但同時亦擁有較高的風險,在測試期間期末淨值雖略低於原策略,但同時其風險亦較小。zh_TW
dc.description.abstractThe investment in stock market is viewed as a game of investor and market in this study. We consider the experience investors participating in the investment game, and apply genetic algorithms to develop investment strategy. Game theory is a bag of analytical tools designed to help us understand the phenomena that we observe when decision-makers interact. We study three different types of investment strategies: day trading, trade strategy and money management. The empirical results show that genetic algorithms are very good at developing investment strategies in training periods, but these strategies may not carry over to testing periods.en_US
DC.subject投資策略zh_TW
DC.subject賽局zh_TW
DC.subject遺傳演算法zh_TW
DC.subjectInvestment Strategyen_US
DC.subjectGameen_US
DC.subjectGenetic Algorithmsen_US
DC.title演化式賽局於投資策略之研究zh_TW
dc.language.isozh-TWzh-TW
DC.titleA Study of Evolutionary Game on Investment Strategyen_US
DC.type博碩士論文zh_TW
DC.typethesisen_US
DC.publisherNational Central Universityen_US

若有論文相關問題,請聯絡國立中央大學圖書館推廣服務組 TEL:(03)422-7151轉57407,或E-mail聯絡  - 隱私權政策聲明