基于小波方法的系统风险多尺度分析

发布时间：2018-01-03 18:35

本文关键词：基于小波方法的系统风险多尺度分析　出处：《兰州商学院》2013年硕士论文　论文类型：学位论文

【摘要】：金融市场的交易者关心的问题之一是风险的测度，由于这些交易者有着不同的投资时间尺度和风险偏好，这使得单一尺度上的风险度量无法满足这些交易者的需求，因此对金融市场的系统风险进行多尺度分析就变得很有必要了。传统方法虽可以通过多建几个模型来进行多尺度分析，但在多尺度分析上小波分析显然更有优势。本文旨在将小波分析和传统的市场模型相结合对金融市场的系统性风险进行多尺度分析，，期待所得的结论能够为投资者做出投资决策提供更具有针对性的信息。本文所采用的小波分析使用MATLAB软件完成。首先运用小波分析研究了上海证券市场中所选个股及所选股票组成的投资组合的系统风险在不同时间尺度上的影响。接下来使用小波方差研究了股票收益率的时间尺度属性，发现小波方差和时间尺度之间表现出近似线性的关系。通过使用小波交叉相关分析上证180指数和所选投资组合收益率之间在不同时间尺度上的关系则表明它们之间的联系随着小波尺度的变化而变化，而且时间尺度越长，交叉相关性越强。对系统性风险的小波分析说明，单个股票和多样化的股票投资组合都具有多尺度性，这也表明通过市场模型度量的β系数即系统性风险随着时间尺度的变化而变化。对于在险价值在不同时间尺度上的分析表明，风险更加集中于数据的高频部分即低的时间尺度上。最后对前述分析进行了扩展，通过计算夏普比率和修正后的夏普比率评估了所选投资组合在不同的时间尺度上的表现。虽然本文所得的结论对于市场中的交易者进行投资决策具有一定的参考价值，但对于系统风险进行多尺度分析的方法仍有改进的余地。这一改进可以从数学方法的角度出发：如使用小波包变换或者提升小波变换，也可以从经济理论的角度出发：如依据相关经济理论对市场模型进行改进，等等。这些都是今后进一步研究的方向。
[Abstract]:One of the concerns of traders in financial markets is the measurement of risk, because these traders have different investment time scales and risk preferences. This makes the risk measurement on a single scale unable to meet the needs of these traders. Therefore, it is necessary to carry on the multi-scale analysis to the system risk of the financial market. Although the traditional method can carry on the multi-scale analysis by building several more models. However, wavelet analysis is obviously more advantageous in multi-scale analysis. This paper aims to combine wavelet analysis with traditional market model to analyze the systemic risk of financial market. Expected conclusions can provide investors with more targeted information to make investment decisions. The wavelet analysis used in this paper is completed by MATLAB software. Firstly, the systematic risk of the selected stocks and the selected stocks in the Shanghai stock market is studied at different time scales by using the wavelet analysis. Then we use wavelet variance to study the time scale attribute of stock return. It is found that there is an approximate linear relationship between wavelet variance and time scale. By using wavelet cross-correlation analysis, the relationship between Shanghai Stock Exchange 180 index and the return rate of selected portfolio on different time scales is shown. The relationship between us changes with the change of wavelet scale. And the longer the time scale, the stronger the cross-correlation. The wavelet analysis of systemic risk shows that single stock and diversified stock portfolio are multi-scale. This also indicates that the 尾 coefficient measured by the market model, namely, the systemic risk changes with the change of the time scale, and the analysis of the value at risk at different time scales shows that. The risk is more concentrated on the high frequency part of the data, that is, the low time scale. Finally, the analysis is extended. The performance of the selected portfolio on different time scales is evaluated by calculating the Sharp ratio and the modified Sharp ratio. Although the conclusions obtained in this paper have certain reference value for traders in the market to make investment decisions. However, there is still room for improvement in the method of multi-scale analysis of system risk, which can be improved from the point of view of mathematical methods, such as using wavelet packet transform or lifting wavelet transform. We can also proceed from the angle of economic theory, such as improving the market model according to the relevant economic theory, and so on.
【学位授予单位】：兰州商学院
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F224;F830.91

【参考文献】