随着中国经济的结构转型以及国际化程度的提高，中国金融市场面临的不确定越来越高，风险暴露的程度日益增加，维持金融系统稳定，防范系统性风险的压力也不断增大。加强对中国金融市场系统性风险的研究，准确度量和识别系统风险，了解系统风险的影响因素和传导机制，对防范系统风险，提高金融系统稳定性具有重要意义。 金融市场的系统性风险测量可以通过资产收益率的分布来刻画。金融市场的收益分布呈现出典型的尖峰厚尾特征，为了控制尾部风险造成的庞大亏损，所以我们应该精确的对风险进行测量以捕获尾部风险。金融市场收益是一种风险收益，反映了投资者对市场的预期，基于风险收益下的系统风险测度会受到市场预期的影响。因此，本文首先使用随机折现因子理论，提出了风险中性条件下的金融系统风险测度即VaR、ES、MES和SES的方法。然后，使用2014-2016年上证指数和恒生指数，分析了中国股票市场和香港股票市场从2014年至2016年的系统风险的变化情况。 本文的实证结果表明，在2015年两个市场都出现市场崩盘的系统风险发生时期，基于随机折现因子理论得到的风险中性系统风险测度明显优于传统的系统风险测度。同时，风险中性的系统风险测度的变化先于市场实际系统风险的变化，即对金融市场的系统性风险具有一定的预警性。这表明风险中性系统风险测度在管理系统风险方面更具有优势。本文的研究结论具有一定的应用价值。

As China‘s economy is still within a structural transformation stage, and its development is involving higher extend of globalization, leading to an increased level of uncertainty on China’s financial market. The pressure comes from whether the systematic risks can be prevented so that the current financial system can be maintained stably. Thus, it is very significant to have a deeper understanding on how to accurately identify and measure the systematic risks, as well as researching on the impact factors and the transmission mechanisms of those risks. The characterization of systematic risks in financial markets can be indicated by the distribution of return on assets (ROA). The distribution has a typical shape of leptokurtosis and fat-tail. In order to prevent the huge losses caused by the fat-tail risks, building a precise risk measurement model becomes more and more critical. Financial market return is a type of risk-return, which reflects the expectations of the investors towards the market. So that systemic risk measurements based on risk-return will be influenced by investor’s expectations. Therefore, this research creatively applies the theory of stochastic discount factor (SDF) to propose the methods for financial system risk measurement under neutral risk conditions for the first time. The methods include VaR, ES, MES and SES. The Shanghai Stock Exchange Index and the Hang Seng Index (2014-2016) are then used as examples to be analyzed on the changes of systemic risks in China and Hong Kong’s stock markets. In conclusion, the research results show that the systematic risk measurements based on the stochastic discount factor theory is significantly better than the traditional risk measurement methods, during the economic downturn in 2015 for both China and Hong Kong’s stock markets. Moreover, the changes of risk-neutral system risk measurements is ahead of the actual system risk in the real market. Thus, it can be served as an early warning to take preventive actions to manage the risks in advance.Overall speaking, the risk-neutral system risk measurements have significant advantages in terms of managing the systematic risks and stabilizing the financial market. The research results can have a wide range of applications.