基于人工神经网络的证券投资基金风格资产轮换策略研究
发布时间：2018-08-22 10:55
【摘要】：随着我国金融市场的日益成熟，国际化进程的加快，证券投资基金规模不断的扩大，风格投资理论也越来越受到广大投资者和金融研究工作者的关注，成为现代投资理论研究的热点问题。 本文选取我国证券市场上90只开放式股票型投资基金为样本，用风格事先划分的方法可以将这90只基金按照其所持股票市值（规模）大小分为小盘型和大盘型，按照其所持股票的加权平均市盈率大小分为价值型和成长型。通过对2001年-2011年小/大盘型、价值/成长型基金收益之差进行分析得出，我国小盘型基金与大盘型基金的收益率表现为相互交替在均值上下波动，，没有一种能够始终保持优势；而价值型基金的市场表现却明显的优于成长型基金，但在个别时段成长型基金还是可以超越价值型基金的。这说明在我国证券投资基金市场上，风格资产的收益具有周期性。 在我国证券市场上风格资产收益存在周期性的前提下，运用现代统计分析方法——人工智能神经网络方法建立风格资产收益预测模型，并对2010年-2011年我国证券投资基金市场上小/大盘风格资产和价值/成长型风格资产的收益表现进行预测，根据预测结果建立了规模风格资产轮换策略和价值/成长型风格资产轮换策略，对比它们与消极的持有单一风格资产的投资策略的业绩发现：在我国证券投资基金市场上实施积极的规模风格资产轮换策略的业绩显著优于消极的持有小/大盘风格资产的投资策略；消极的持有价值型风格资产的投资策略的业绩明显优于实施积极的价值/成长型风格资产轮换策略及消极的持有成长型风格资产的投资策略。这说明，在我国证券投资基金市场上适合实施积极的风格资产轮换策略，而不适合实施积极的价值/成长型风格资产轮换策略。
[Abstract]:With the maturation of our financial market and the acceleration of the internationalization process, the scale of the securities investment funds is constantly expanding, and the theory of style investment has been paid more and more attention by investors and financial researchers. It has become a hot issue in the research of modern investment theory. In this paper, 90 open-end equity investment funds in China's securities market are selected as samples, and the 90 funds can be divided into small market type and large market type according to their stock market value (scale) by the method of style pre-division. It is divided into value type and growth type according to the weighted average price-earnings ratio of its stock. Through the analysis of the difference between the small / large market type, the value / growth fund income from 2001 to 2011, it is concluded that the return rate of the small-cap fund and the large-cap fund in China fluctuates alternately between the average value and the average value, and none of them can always maintain the advantage. But the market performance of the value fund is obviously better than that of the growth fund, but the growth fund can still surpass the value fund in some time. This shows that in China's securities investment fund market, the return of style assets has periodicity. On the premise of the periodicity of the return of style assets in the securities market of our country, the forecasting model of the income of style assets is established by using the modern statistical analysis method-artificial intelligence neural network method. And forecast the income performance of small / large market style assets and value / growth style assets in China's securities investment fund market from 2010 to 2011. Based on the prediction results, the scale style asset rotation strategy and the value / growth style asset rotation strategy are established. Comparing their performance with negative investment strategies with single style assets, it is found that the performance of positive scale style asset rotation strategy in China's securities investment fund market is significantly better than that of negative holding small / large market. Investment strategy of style assets; The negative investment strategy of value style assets is obviously superior to the positive value / growth style asset rotation strategy and the negative investment strategy of holding the growth style assets. This shows that it is suitable to implement the positive style asset rotation strategy in China's securities investment fund market, but not the positive value / growth style asset rotation strategy.
【学位授予单位】：兰州商学院
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F832.51;TP18
[Abstract]:With the maturation of our financial market and the acceleration of the internationalization process, the scale of the securities investment funds is constantly expanding, and the theory of style investment has been paid more and more attention by investors and financial researchers. It has become a hot issue in the research of modern investment theory. In this paper, 90 open-end equity investment funds in China's securities market are selected as samples, and the 90 funds can be divided into small market type and large market type according to their stock market value (scale) by the method of style pre-division. It is divided into value type and growth type according to the weighted average price-earnings ratio of its stock. Through the analysis of the difference between the small / large market type, the value / growth fund income from 2001 to 2011, it is concluded that the return rate of the small-cap fund and the large-cap fund in China fluctuates alternately between the average value and the average value, and none of them can always maintain the advantage. But the market performance of the value fund is obviously better than that of the growth fund, but the growth fund can still surpass the value fund in some time. This shows that in China's securities investment fund market, the return of style assets has periodicity. On the premise of the periodicity of the return of style assets in the securities market of our country, the forecasting model of the income of style assets is established by using the modern statistical analysis method-artificial intelligence neural network method. And forecast the income performance of small / large market style assets and value / growth style assets in China's securities investment fund market from 2010 to 2011. Based on the prediction results, the scale style asset rotation strategy and the value / growth style asset rotation strategy are established. Comparing their performance with negative investment strategies with single style assets, it is found that the performance of positive scale style asset rotation strategy in China's securities investment fund market is significantly better than that of negative holding small / large market. Investment strategy of style assets; The negative investment strategy of value style assets is obviously superior to the positive value / growth style asset rotation strategy and the negative investment strategy of holding the growth style assets. This shows that it is suitable to implement the positive style asset rotation strategy in China's securities investment fund market, but not the positive value / growth style asset rotation strategy.
【学位授予单位】：兰州商学院
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F832.51;TP18
【参考文献】
相关期刊论文 前5条
1 邓可斌;唐s
本文编号：2196880
本文链接：https://www.wllwen.com/guanlilunwen/zhqtouz/2196880.html
最近更新
教材专著
