Mid-term inspection of students' scientific research projects at Hangzhou University of electronic science and technology-Complex Network Modeling Analysis of stock Networks

Project name: Complex Network Modeling and Analysis of stock Network

Project category: Combined with scientific research

Approval time: 2011.1

Planned completion time: 2011.12

Project objectives: 1. Develop software that can generate complex stock Networks Based on Datasets

2. Write a scientific research paper on Complex Network Modeling and Analysis of stock Networks

Project Progress:

1. The project team members learned the related concepts and basic knowledge of graph theory, had a clear understanding of the origin of Complexity science, and had a preliminary understanding of the basic characteristics of complex networks.

2. Data Extraction

Because our existing data includes treasury bonds, funds, futures, and stocks, extracting the required stock data is a necessary preliminary step for the startup project, currently, we use C programming to extract the required stock data from the file, including 852 Shanghai Stock and 792 Shenzhen stock.

3. Construction of the securities market network

To better measure the correlation between stocks, we have introduced a scale-free Weighted Network for the correlation between the extracted 852 Shanghai Stock and 792 Shenzhen stock, the node indicates the stock, and the Edge Weight Coefficient CIJ indicates the association strength between the node and the node. To give a simple association between stocks, we define a threshold θ. If | CIJ | <θ, CIJ = 0; otherwise, CIJ = 1, in this way, we can build a 0-1 Unauthorized network. For this network, we can analyze its degree distribution, clustering coefficient, average path length, and other structural characteristics.

Next work plan

We will start from another angle, such as transaction volume, and then build another complex stock network, which allows us to analyze and better understand the association between stocks from different angles; based on the constructed complex network, we will further analyze the network's non-scale, small-world characteristics and other topological features, understand its physical meaning, and explain the characteristics of the network, analyze the internal mechanism of its evolution. We will use pajek software to plot the complex network and draw out the topology characteristics of the complex network on the MATLAB platform. Finally, we will discuss the method to study the optimal portfolio model using the complex network theory, explore the feasibility of this achievement in the real stock market. At the end of this semester, our team will discuss and clarify their respective tasks once again, and will make full use of the summer period to complete the work.

 Use of scientific research funds

Copy printing fee: 68 RMB

Library fee: 300 yuan

Transportation fee: 300 yuan