學術產出-國科會研究計畫

文章檢視/開啟

書目匯出

Google ScholarTM

政大圖書館

引文資訊

TAIR相關學術產出

題名 神經網路在選擇權定價上之應用
作者 陳威光;蔡瑞煌
關鍵詞 類神經網路;選擇權定價;亞式選擇權;倒傳遞類神經網路
Neural network;Option pricing;Asian option;Back propagation neural network
日期 1999
上傳時間 18-四月-2007 16:34:53 (UTC+8)
出版社 臺北市:國立政治大學金融系
摘要 應用類神經網路的技術在財務金融領域是最近幾年來熱門的課題,如同Kuan and White(1994)所言:神經網路是一種非線性參數模型,而網路系統的學習過程好比進行模型參數之估計。本文採用最常被應用的倒傳遞網路系統,以及理解神經網路兩系統來做為亞式選擇權的評價。隨著特殊選擇權產品的快速發展,如亞式選擇權(或稱平均式選擇權),這些選擇權的定價公式對理論界或實務界也愈來愈有其迫切性。然而最為市場採用的算術平均選擇權,因其分配不是對數常態分配,因此並沒有標準公式解。大多是以數值的方法逼近。本研究的目的在於利用神經網路非線性的學習特性來評價這些選擇權。本文採用神經網路的學習系統來評價亞式選擇權。本文結果發現網路學習效果具有相當不錯的準確性。
It has been a hot issue to utilize the technique of artificial Neural Network (ANN) to the financial markets such as stocks, and futures markets. The purpose of this paper is to apply the Back Propagation Neural Network (BP) and Reasoning Neural Network (RN) to the valuation of Asia option. The option markets have grown dramatically since Black-Scholes (1973) derived their famous option pricing model. The markets for the exotic products especial for Asia options (average rate option) have grown rapidly for the past decade. There is no closed-form solution for the arithmetic average option since the distribution of the sum of the stock price is not lognormal. The propose of the paper is to utilize the BP Neural Network as well as RN Neural Network to price the Asia option. It is shown that artificial neural networks can be successfully employed to approximate formulas for the Asian option, for which analytic formulas cannot derived.
描述 核定金額:238500元
資料類型 report
dc.coverage.temporal 計畫年度:88 起迄日期:19980801~20000430en_US
dc.creator (作者) 陳威光;蔡瑞煌zh_TW
dc.date (日期) 1999en_US
dc.date.accessioned 18-四月-2007 16:34:53 (UTC+8)en_US
dc.date.accessioned 8-九月-2008 16:41:58 (UTC+8)-
dc.date.available 18-四月-2007 16:34:53 (UTC+8)en_US
dc.date.available 8-九月-2008 16:41:58 (UTC+8)-
dc.date.issued (上傳時間) 18-四月-2007 16:34:53 (UTC+8)en_US
dc.identifier (其他 識別碼) 882416H004013.pdfen_US
dc.identifier.uri (URI) http://tair.lib.ntu.edu.tw:8000/123456789/3730en_US
dc.identifier.uri (URI) https://nccur.lib.nccu.edu.tw/handle/140.119/3730-
dc.description (描述) 核定金額:238500元en_US
dc.description.abstract (摘要) 應用類神經網路的技術在財務金融領域是最近幾年來熱門的課題,如同Kuan and White(1994)所言:神經網路是一種非線性參數模型,而網路系統的學習過程好比進行模型參數之估計。本文採用最常被應用的倒傳遞網路系統,以及理解神經網路兩系統來做為亞式選擇權的評價。隨著特殊選擇權產品的快速發展,如亞式選擇權(或稱平均式選擇權),這些選擇權的定價公式對理論界或實務界也愈來愈有其迫切性。然而最為市場採用的算術平均選擇權,因其分配不是對數常態分配,因此並沒有標準公式解。大多是以數值的方法逼近。本研究的目的在於利用神經網路非線性的學習特性來評價這些選擇權。本文採用神經網路的學習系統來評價亞式選擇權。本文結果發現網路學習效果具有相當不錯的準確性。-
dc.description.abstract (摘要) It has been a hot issue to utilize the technique of artificial Neural Network (ANN) to the financial markets such as stocks, and futures markets. The purpose of this paper is to apply the Back Propagation Neural Network (BP) and Reasoning Neural Network (RN) to the valuation of Asia option. The option markets have grown dramatically since Black-Scholes (1973) derived their famous option pricing model. The markets for the exotic products especial for Asia options (average rate option) have grown rapidly for the past decade. There is no closed-form solution for the arithmetic average option since the distribution of the sum of the stock price is not lognormal. The propose of the paper is to utilize the BP Neural Network as well as RN Neural Network to price the Asia option. It is shown that artificial neural networks can be successfully employed to approximate formulas for the Asian option, for which analytic formulas cannot derived.-
dc.format applicaiton/pdfen_US
dc.format.extent bytesen_US
dc.format.extent 1539452 bytesen_US
dc.format.extent 1539452 bytes-
dc.format.extent 928 bytes-
dc.format.mimetype application/pdfen_US
dc.format.mimetype application/pdfen_US
dc.format.mimetype application/pdf-
dc.format.mimetype text/plain-
dc.language zh-TWen_US
dc.language.iso zh-TWen_US
dc.publisher (出版社) 臺北市:國立政治大學金融系en_US
dc.rights (權利) 行政院國家科學委員會en_US
dc.subject (關鍵詞) 類神經網路;選擇權定價;亞式選擇權;倒傳遞類神經網路-
dc.subject (關鍵詞) Neural network;Option pricing;Asian option;Back propagation neural network-
dc.title (題名) 神經網路在選擇權定價上之應用zh_TW
dc.type (資料類型) reporten