Maximal spanning trees, asset graphs and random matrix denoising in the analysis of dynamics of financial networks

Heimo, Tapio; Kaski, Kimmo; Saramäki, Jari

doi:10.1016/j.physa.2008.10.007

Cited by 31 publications

(31 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Pan and Sinha [445] investigated the stock price fluctuations in the National Stock Exchange (NSE) of India, and found that: (i) stocks in emerging markets are more correlated than in developed ones and (ii) the Indian market evolved into clusters corresponding to business sectors. The clustering property has also been investigated in the NYSE [446][447][448][449]. Bolgorian and Raei [450] considered thresholding in correlations between daily stock prices to build a network.…”

Section: Financial Marketmentioning

confidence: 99%

Analyzing and modeling real-world phenomena with complex networks: a survey of applications

Costa

Oliveira

Travieso

et al. 2011

Advances in Physics

659

357

View full text Add to dashboard Cite

Section: Financial Marketmentioning

confidence: 99%

Analyzing and modeling real-world phenomena with complex networks: a survey of applications

Costa

Oliveira

Travieso

et al. 2011

Advances in Physics

659

357

View full text Add to dashboard Cite

“…Spanning tree is a particular type of graph without forming any loop, so a maximal spanning tree connects all the N nodes of a network with N − 1 links, as a result, the sum of the link weights is maximized. 27 Maximal spanning tree can be extracted by Prim's Algorithm. Corresponding maximal spanning trees of industry network are shown in Fig.…”

Section: Maximal Spanning Treementioning

confidence: 99%

Effects of financial crisis on the industry sector of Chinese stock market — from a perspective of complex network

et al. 2014

View full text Add to dashboard Cite

In this paper, we use mutual information to measure the statistical interdependence between 23 industry sectors of Shanghai stock market and construct corresponding correlation network to analyze the shock of 2008 financial crisis on industry sectors. The obtained meaningful facts are as follows. First, such crisis has only a limited impact on leading industries such as Manufacturing, Commercial trade and Machinery & Equipment, which still play an important role in Chinese economy. Second, the crisis badly attacks China's export industries like Electronics, Wood & Furniture and Textile & Clothing. The damage further hurts other industries, and then export industries' influence becomes larger. Third, the crisis adversely impacts the import industries like Petrochemical, Metal & Nonmetal and Pharmaceutical Biotechnology. While due to the stimulation of macroeconomic policies, the influence of crisis on import industries is limited. Similarly, due to relatively strict capital control and the macroeconomic policies stimulating the domestic demand, those industries like Construction, Real Estate and Financial Services are slightly wounded. All these findings suggest that Chinese government should transform from the external demand to the domestic consumption to sustain economic growth.

show abstract

“…RMT is often used to analyze the information in the stock correlation matrix . Laloux et al analyzed the correlation matrix for S&P 500 through RMT method and found that about 94% of the eigenvalues of correlation matrix, which correspond to the noise in the stock market, are in the theoretical area of eigenvalues of random matrix, and this method could be used in the risk management of investment portfolio .…”

Section: Introductionmentioning

confidence: 99%

“…As the natural relation between RMT and complex network, there are some papers to detect the structure of the finance network through RMT . Eom et al reconstructed the stock network according to the eigenvalues reflected.…”

Section: Introductionmentioning

confidence: 99%

The network structure of Chinese finance market through the method of complex network and random matrix theory

Huang²,

Cao

et al. 2018

Concurrency and Computation

View full text Add to dashboard Cite

Summary As we know, economic policy has a significant influence on the stock market. So, it should be interesting to detect the relation between the correlation structure of the stock market and economic policy. To achieve this target, the correlation matrix is analyzed through component analysis method first, to detect the economic implication of the components of correlation matrix. It is found that the biggest driving force of market behavior affects the whole stock market instead of individual stock or plate and the evolution of important plates of the second component relates closely to economic policies. Then, ordinary networks, denoising networks, and noise networks are constructed for different periods and distinction among the structures of different networks in the same period is analyzed. The result suggests that the economic factors corresponding to the first and second components make the stock market be sensitive and lead to a gathering effect in ordinary and denoising networks. At last, based on the inferences from this paper, the formation of power law in the stock network and the influence of crisis on the structure of stock network are discussed.

show abstract

Maximal spanning trees, asset graphs and random matrix denoising in the analysis of dynamics of financial networks

Cited by 31 publications

References 26 publications

Analyzing and modeling real-world phenomena with complex networks: a survey of applications

Analyzing and modeling real-world phenomena with complex networks: a survey of applications

Effects of financial crisis on the industry sector of Chinese stock market — from a perspective of complex network

The network structure of Chinese finance market through the method of complex network and random matrix theory

Contact Info

Product

Resources

About