Correlation network analysis for multi-dimensional data in stocks market

Kazemilari, Mansooreh; Djauhari, Maman A.

doi:10.1016/j.physa.2015.02.052

Cited by 41 publications

(26 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The off-diagonal elements of VC vary between 0 to +1, where 0 indicates that each OHLC price of currency p i is uncorrelated to each OHLC prices of p j , that is, Cov ij = 0. ρV ij is equal to +1 when the configurations of OHLC prices of two currencies are same. See also Escoufier (2006), Kazemilari and Djauhari (2015) and Smilde et al (2009) for more properties of ρV.…”

Section: Similarity Measure In the Multivariate Setting Of O H L Anmentioning

confidence: 99%

“…In all previous models of network analysis in FOREX market, the behavior of closing price has been studied, in which each return is represented as univariate time series. However, in financial market activities, currencies and stocks are actually represented as Multivariate Time Series (MTS) of Opening price, Highest price, Lowest price and Closing price (OHLC) (Kazemilari and Djauhari 2015;Gan and Djauhari 2015). There have been no studies done on the network analysis of currencies in relation to the crisis in multivariate setting.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Topological Network Analysis Based on Dissimilarity Measure of Multivariate Time Series Evolution in the Subprime Crisis

Kazemilari

Mohamadi

2018

IJFS

Self Cite

View full text Add to dashboard Cite

Correlation network based on similarity is the common approach in financial network analyses where the Minimal Spanning Tree (MST) is used to filter the important information contained in the network. In this paper, by considering a distance matrix based on dissimilarities among multivariate time series of currency, a topological network was analyzed. A topological network can explain to what extent two or more multi-dimensional currency structures are different from each other. For this purpose, we examined the topological network of currency market from 2005 to 2011 in terms of the subprime crisis. After that, the multivariate time series evolution of MSTs were analyzed in terms of the structural changes for three periods (before, during, and after the crisis). Moreover, since the clusters of currencies in network analysis are due to regional factors, by considering each region, which is composed of a number of currencies, as an element on the financial system, we attempted to determine how a region interacts with the other regions in crisis periods. This motivated us to introduce a region-based network analysis of currencies. Since each region consisted of a different number of currencies compared to the others, the appropriate network analysis was in multivariate setting. Finally, the applications of the method were presented with the situation of a currencies crisis behavior. The results indicate significant changes in the topological structures of MSTs when their properties are compared to each other. Int. J. Financial Stud. 2018, 6, 47 2 of 16 management of assets, is considered to be an important topic in this regard. From a statistical physics perspective, correlation analysis between financial objects has developed in various approaches, such as the multi-fractal analysis theory, random matrix theory, and correlation network-based approaches (e.g., the planar maximally filtered graph, the minimum spanning tree (MST), and the correlation threshold approaches). Particularly, as Mantegna (1999) initially proposed the network analysis tool of MST in the US stock market, the correlation network was applied for quantifying the interconnections in many financial markets, such as the commodity markets, foreign exchange (FOREX, FX) market, stock market, and equity markets.In financial markets, topological network analysis is a technical method that provides efficient tools to interpret market properties and structures (Mantegna 1999;Mantegna and Stanley 2000;Tumminello et al. 2005; Djauhari and Lee 2014). Network topology refers to the physical layout of a network. This approach has been investigated seriously by econophysicists to understand the structure of interrelated variables via network. It defines the way that different nodes are placed and are interconnected with each other. Alternately, technical topology networks describe how the data is transferred between these nodes. This is achieved by employing the well-known correlation network analysis approach. On the other hand, correlation network is usua...

show abstract

Section: Similarity Measure In the Multivariate Setting Of O H L Anmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Topological Network Analysis Based on Dissimilarity Measure of Multivariate Time Series Evolution in the Subprime Crisis

Kazemilari

Mohamadi

2018

IJFS

Self Cite

View full text Add to dashboard Cite

show abstract

“…RN Mantegna [31] mapped the financial market as a network, the vertices of which are stocks and the edges between vertices are the relationships of shares. Kazemilari et al [32] constructed a multivariate correlation network of stocks where each of them was represented by a multivariate time series, and vector correlation was used to measure the similarity among multivariate time series in the stock network.…”

Section: Introductionmentioning

confidence: 99%

Construction of Complex Network with Multiple Time Series Relevance

Huang

Wang³

et al. 2018

Information

View full text Add to dashboard Cite

Multivariate time series data, which comprise a set of ordered observations for multiple variables, are pervasively generated in weather conditions, traffic, financial stocks, etc. Therefore, it is of great significance to analyze the correlation between multiple time series. Financial stocks generate a significant amount of multivariate time series data that can be used to build networks that reflect market behavior. However, traditional commercial complex networks cannot fully utilize the multiple attributes of stocks and redundant filter relationships and reveal a more authentic financial stock market. We propose a fusion similarity of multiple time series and construct a threshold network with similarity. Furthermore, we define the connectivity efficiency to choose the best threshold, establishing a high connectivity efficiency network with the optimal network threshold. By searching the central node in the threshold network, we have found that the network center nodes constructed by our proposed method have a more comprehensive industry coverage than the traditional techniques to build the systems, and this also proves the superiority of this method.

show abstract

“…According to given threshold, Tse Chi K [3] established stock correlation network with winner-win standard. Mansooreh Kazemilari [4] introduced the concept of vector relation, and assumed each stock is a multivariate time series. Then, he established stock correlation network, analyzed the structure of minimum dendrogram.…”

Section: Introductionmentioning

confidence: 99%

Complex character Analysis of Shenzhen 100 index Stock Market Network

Tu¹,

Yang²

2015

Proceedings of the 3rd International Conference on Mechatronics and Industrial Informatics

View full text Add to dashboard Cite

By the data of these stocks ， daily closing prizing, volume, and return, we built three kinds of stock market network which based on the concept of complex network. First of all, we calculated the correlation index between stock nodes, which were expressed by a correlation coefficient matrix. According to kernel density distribution of daily closing prizing, volume, and return ' s correlation coefficient ,we determined their threshold and established the stock correlation networks.Then we analyzed character of small world for these stock correlation networks. Finally, by indicator for degree and betweenness centrality, we analyzed centrality for trading volume stock market network, which have strongest character of small world comparing with other networks.

show abstract

Correlation network analysis for multi-dimensional data in stocks market

Cited by 41 publications

References 56 publications

Topological Network Analysis Based on Dissimilarity Measure of Multivariate Time Series Evolution in the Subprime Crisis

Topological Network Analysis Based on Dissimilarity Measure of Multivariate Time Series Evolution in the Subprime Crisis

Construction of Complex Network with Multiple Time Series Relevance

Complex character Analysis of Shenzhen 100 index Stock Market Network

Contact Info

Product

Resources

About