| 信息技术企业股票关联网络的风险传染研究 |
| Cascading Failure Risk Research on Stock Associative Network of Information Technology Enterprises |
| 摘要点击 259 全文点击 0 投稿时间：2020-05-11 修订日期：2022-12-01 |
| 查看/发表评论 下载PDF阅读器 |
| 中文关键词 时间序列;最长公共子序列;关联网络;级联失效;风险传染 |
| 英文关键词 Time series; Longest Common Subsequence; Similarity Network ; Cascading Failure; Risk Contagion |
| 基金项目 上海市哲学社会科学规划一般项目, 信息技术企业与银行相依网络的风险传染研究（2019BGL003）,主持人:张晓黎. |
| 作者 | 单位 | 邮编 | | 张晓黎* | 上海对外经贸大学 统计与信息学院 | 201102 |
|
| 中文摘要 |
| 基于整数编码模式最长公共子序列相似度算法,计算信息技术企业间的股票价格时间序列相似度。相似度阈值超过0.84的企业间建立股票关联网络。企业间共形成了16个团体,其中最大的团体有145家企业。单个到多个企业受外部攻击后,关系强度波动和关系双方资产受损,风险沿路径传染的结果：按企业资产受外部攻击后的风险传染时长、倒闭企业数量明显高于按企业在网络位置的。风险传染破坏力与外部攻击强度分段有关,在外部攻击强度为1到10级时攀升最快。风险传染破坏力在受攻击企业数量小于80时攀升最快,超过此阈值后趋缓。 |
| 英文摘要 |
| Based on the longest common subsequence similarity algorithm of integer coding mode, the stock price time series similarity among information technology enterprises is calculated. When the threshold of similarity is more than 0.84, relationships are established to form the stock similarity network. There are 16 groups in the stock similarity network, among which the largest group has 145 companies. Single to multiple enterprises are deliberately attacked, resulting in the fluctuation of relationship strength, the assets damage on the both sides, and risk contagion along the path. The contagion duration and the number of bankrupt enterprises caused by the external attacks on enterprises with large assets are significantly higher than enterprises on the good locations in the network. The contagion duration is related to the intensity of external attack when the level is between 1 and 10. The destructive power of risk contagion climbs fastest while the number of the attacked enterprises is below 80, and alleviates after exceeding the attack intensity threshold. |
| 关闭 |
