基于有向有限穿越可视图的时间序列伴生网络

发布时间：2018-03-16 17:25

本文选题：时间序列　切入点：复杂网络　出处：《系统工程学报》2017年02期 　论文类型：期刊论文

【摘要】：提出两种新的可视图建网方法(有限穿越水平可视图和有向有限穿越可视图),有限穿越水平可视图综合了水平可视图的算法简便性以及有限穿越可视图的抗噪声能力,而有向有限穿越可视图则突破了原有算法的无向性,考虑了时间的不可逆性.与传统可视图的三种模型进行对比,验证了两种算法的有效性,最后将模型应用到股市进行了实证分析.
[Abstract]:In this paper, two new visual graph construction methods, namely, finite traversing horizontal visual graph and directed finite traversing visual graph are proposed. The finite traversing horizontal visual graph combines the algorithm simplicity of horizontal visual graph and the anti-noise ability of finite traversing visual graph. The directional finite traversing visual plan breaks through the irreversibility of the original algorithm and considers the irreversibility of time. Compared with the three models of traditional visual graph, the validity of the two algorithms is verified. Finally, the model is applied to the stock market for empirical analysis.
【作者单位】：暨南大学管理学院;
【基金】：广东省科技发展专项资金资助项目(2016B040401003);广东省科技发展专项资金资助项目(2016A010101016) 广东省重大科技专项资助项目(2016B010124008)
【分类号】：F830.91;O157.5

【相似文献】