| dc.contributor.advisor | 黃泓智 | zh_TW |
| dc.contributor.advisor | Huang, Hong-Chih | en_US |
| dc.contributor.author (Authors) | 黃勝彥 | zh_TW |
| dc.contributor.author (Authors) | Huang, Sheng-Yan | en_US |
| dc.creator (作者) | 黃勝彥 | zh_TW |
| dc.creator (作者) | Huang, Sheng-Yan | en_US |
| dc.date (日期) | 2020 | en_US |
| dc.date.accessioned | 3-Aug-2020 17:41:33 (UTC+8) | - |
| dc.date.available | 3-Aug-2020 17:41:33 (UTC+8) | - |
| dc.date.issued (上傳時間) | 3-Aug-2020 17:41:33 (UTC+8) | - |
| dc.identifier (Other Identifiers) | G0107358012 | en_US |
| dc.identifier.uri (URI) | http://nccur.lib.nccu.edu.tw/handle/140.119/131007 | - |
| dc.description (描述) | 碩士 | zh_TW |
| dc.description (描述) | 國立政治大學 | zh_TW |
| dc.description (描述) | 風險管理與保險學系 | zh_TW |
| dc.description (描述) | 107358012 | zh_TW |
| dc.description.abstract (摘要) | 本研究主要是以長短期記憶模型（LSTM）進行基金報酬率預測，搭配波動度控制的方法，建構穩健的投資組合。為了達成風險分散的目的，本論文將股票型基金與債券型基金分別處理，其中，股票型基金劃分為六大市場，包含日本、美國、新興市場、歐洲、亞太不含日本與台灣，而債券型基金劃分為三大類別，包含投資級別債券、高收益債券以及新興市場債券，於資產配置時，從各大地區與類別挑選出預期表現較佳之基金，達成風險分散之目的。在波動度控制的部分，本文以下方標準差做為波動度的衡量，並嘗試以固定波動度與變動波動度的方法進行資產配置，最終比較其結果之差異。實證結果發現，透過每月檢視投資組合的風險，變動波動度控制能夠迅速反應市場狀況，且較為保守，整體績效優於固定波動度控制。 | zh_TW |
| dc.description.abstract (摘要) | The purpose of this study is to use LSTM model to predict the return of mutual fund, and build a stable portfolio. In order to achieve the purpose of risk diversification, this study treats bond fund and stock fund separately. Moreover, stock funds are divided into six major markets, and bond funds are divided into three major categories. The final portfolio will include funds from each category in order to diversify the risk. This study uses two volatility control methods to determine the asset allocation, including fixed volatility control and variable volatility control. The empirical results find that through monthly review of portfolio risk, variable volatility control method can quickly reflect market conditions, and therefore the overall performance is better than fixed volatility control. | en_US |
| dc.description.tableofcontents | 第一章 緒論 1第一節 研究動機與背景 1第二節 研究目的 2第三節 研究流程 3第二章 文獻探討 4第一節 深度學習之文獻探討 4第二節 深度學習與股價預測相關之文獻探討 5第三節 資產配置之文獻探討 5第三章 研究方法 7第一節 研究架構 7第二節 基金分群 8第三節 動能投資策略（Momentum Investment Strategy） 11第四節 均線策略（Moving Average Strategy） 11第五節 長短期記憶模型（Long Short-Term Memory, LSTM） 12第六節 資產配置策略 20第七節 績效指標說明 26第四章 實證結果 27第一節 實證分析樣本來源 27第二節 固定波動度控制 27第三節 變動波動度控制─每半年配置股債權重 30第四節 變動波動度控制─每月配置股債權重 33第五章 結論與未來方向建議 36參考文獻 38 | zh_TW |
| dc.source.uri (資料來源) | http://thesis.lib.nccu.edu.tw/record/#G0107358012 | en_US |
| dc.subject (關鍵詞) | 基金分群 | zh_TW |
| dc.subject (關鍵詞) | 長短期記憶模型 | zh_TW |
| dc.subject (關鍵詞) | 波動度控制 | zh_TW |
| dc.subject (關鍵詞) | 最適化權重 | zh_TW |
| dc.subject (關鍵詞) | Mutual fund | en_US |
| dc.subject (關鍵詞) | LSTM | en_US |
| dc.subject (關鍵詞) | Volatility control | en_US |
| dc.subject (關鍵詞) | Optimal asset allocation | en_US |
| dc.title (題名) | 利用深度學習模型建構基金最適資產配置 | zh_TW |
| dc.title (題名) | Using Deep Learning Model to Construct The Optimal Asset Allocation In Mutual Fund | en_US |
| dc.type (資料類型) | thesis | en_US |
| dc.relation.reference (參考文獻) | 1.Abe Masaya and Nakayama Hideki（2018）.“Deep learning for forecasting stock returns in the cross-section,” Pacific-Asia Conference on Knowledge Discovery and Data Mining. 273-284.2.Minh Dang Lien, Sadeghi-Niaraki Abolghasem, Huy Huynh Duc, Min Kyungbok and Moon Hyeonjoon（2018）. “Deep learning approach for short-term stock trends prediction based on two-stream gated recurrent unit network,” IEEE Access, 55392-55404.3.A.Connel and M.Hodgson（2017）. “Managing investment outcomes with volatility control.”4.Nelson David MQ., Pereira Adriano C.M. and de Oliveira Renato A.（2017）. “Stock market`s price movement prediction with LSTM neural networks,” 2017 International joint conference on neural networks (IJCNN), 1419-1426.5.Xiong Ruoxuan, Nichols Eric P. and Shen Yuan（2017）. “Deep learning stock volatility with google domestic trends.”6.Chong Eunsuk, Han Chulwoo, and Park Frank C（2017）.“Deep learning networks for stock market analysis and prediction: Methodology, data representations, and case studies,” Expert Systems with Applications, 187-205.7.Roondiwala Murtaza, Patel Harshal and Varma Shraddha（2017）. “Predicting stock prices using LSTM,” International Journal of Science and Research (IJSR), 1754-1756.8.Hansson Magnus（2017）.“On stock return prediction with LSTM networks.”9.Keller Wouter J., Butler Adam and Kipnis Ilya（2015）.“ Momentum and Markowitz: a golden combination.”10.Harry M. Markowitz（2010）. “Portfolio theory: as I still see it,” Annu. Rev. Financ. Econ, 1-23.11.Hochreiter Sepp and Schmidhuber Jürgen（1997）. “Long short-term memory,” Neural computation, 1735-1780.12.Kimoto Takashi, Asakawa Kazuo, Yoda Morio and Takeok Masakazu（1990）. “Stock market prediction system with modular neural networks,” 1990 IJCNN international joint conference on neural networks, 1-6.13.Levy Haim and Sarnat Marshall（1970）. “International diversification of investment portfolios,” The American Economic Review, 668-675.14.Grubel, Herbert G.（1968）. “Internationally diversified portfolios: welfare gains and capital flows,” The American Economic Review, 1299-1314. | zh_TW |
| dc.identifier.doi (DOI) | 10.6814/NCCU202000778 | en_US |
