论文标题
中国市场崩溃的股票网络不足
The illiquidity network of stocks in China's market crash
论文作者
论文摘要
中国股票市场在2015年突然发生了崩溃，其市场价值的三分之一超过三分之一。鉴于其与频率的恐惧和良好的解决方案的关联，股票的流动性不足可能为理解甚至发出市场崩溃提供了有希望的观点。在这项研究中，通过将股票与流动性不足的共同连接起来，建立了一个不流动性网络来建模市场。与无灾难的日子相比，在崩溃日，由于较重但更均匀的流动性依赖性促进突然崩溃的差异依赖性，市场更加密集。流动性网络中的关键股票，特别是金融部门的关键库存，是针对检查的，因为它们在积累和传递流动性损失方面具有关键作用。市场崩溃中股票的级联失败是从小度的传播到通常位于流动性网络核心的高度的高度的传播中，然后返回到外围。通过计算前五天随机失败的日子，实现了早期信号，以成功预测撞车日的一半以上，尤其是在早期阶段的连续几天。 Granger因果关系网络和随机网络的其他证据进一步证明了信号的鲁棒性。我们的结果可以帮助市场从业人员（例如监管机构）检测并防止事先发生坠机风险。
The Chinese stock market experienced an abrupt crash in 2015, and over one-third of its market value evaporated. Given its associations with fear and the fine resolution with respect to frequency, the illiquidity of stocks may offer a promising perspective for understanding and even signaling a market crash. In this study, by connecting stocks with illiquidity comovements, an illiquidity network is established to model the market. Compared to noncrash days, on crash days, the market is more densely connected due to heavier but more homogeneous illiquidity dependencies that facilitate abrupt collapses. Critical stocks in the illiquidity network, particularly those in the finance sector, are targeted for inspection because of their crucial roles in accumulating and passing on illiquidity losses. The cascading failures of stocks in market crashes are profiled as disseminating from small degrees to high degrees that are usually located in the core of the illiquidity network and then back to the periphery. By counting the days with random failures in the previous five days, an early signal is implemented to successfully predict more than half of the crash days, especially consecutive days in the early phase. Additional evidence from both the Granger causality network and the random network further testifies to the robustness of the signal. Our results could help market practitioners such as regulators detect and prevent the risk of crashes in advance.
