Purpose

The purpose of the research is to assess the risk of the financial market in the digital economy through the quantitative analysis model in the big data era. It is a big challenge for the government to carry out financial market risk management in the big data era.

Design/methodology/approach

In this study, a generalized autoregressive conditional heteroskedasticity-vector autoregression (GARCH-VaR) model is constructed to analyze the big data financial market in the digital economy. Additionally, the correlation test and stationarity test are carried out to construct the best fit model and get the corresponding VaR value.

Findings

Owing to the conditional heteroscedasticity, the index return series shows the leptokurtic and fat tail phenomenon. According to the AIC (Akaike information criterion), the fitting degree of the GARCH model is measured. The AIC value difference of the models under the three distributions is not obvious, and the differences between them can be ignored.

Originality/value

Using the GARCH-VaR model can better measure and predict the risk of the big data finance market and provide a reliable and quantitative basis for the current technology-driven regulation in the digital economy.

中文翻译：

---

大数据，大挑战：数字经济下金融市场的风险管理

目的

研究目的是通过大数据时代的量化分析模型评估数字经济下金融市场的风险。政府在大数据时代开展金融市场风险管理是一个很大的挑战。

设计/方法/方法

本研究构建了广义自回归条件异方差向量自回归（GARCH-VaR）模型来分析数字经济中的大数据金融市场。此外，进行相关性检验和平稳性检验，构建最佳拟合模型，得到相应的VaR值。

发现

由于条件异方差，指数收益序列呈现出尖峰和肥尾现象。根据AIC（Akaike信息准则），衡量GARCH模型的拟合度。三种分布下模型的AIC值差异不明显，可以忽略它们之间的差异。

原创性/价值

使用 GARCH-VaR 模型可以更好地衡量和预测大数据金融市场的风险，为当前数字经济中技术驱动的监管提供可靠的量化依据。