Authors:
Zahra Hatami
1
;
Hesham Ali
1
;
David Volkman
2
and
Prasad Chetti
3
Affiliations:
1
College of Information Science & Technology, University of Nebraska at Omaha,Omaha, U.S.A.
;
2
College of Business Administration, University of Nebraska at Omaha,Omaha, U.S.A.
;
3
School of Computer Science & Information Systems,Northwest Missouri State University, Maryville, U.S.A.
Keyword(s):
Financial Markets, Population Analysis, Network Models.
Abstract:
With the availability of massive data sets associated with stock markets, we now have opportunities to apply newly developed big data techniques and data-driven methodologies to analyze these complicated markets. Correlation network analysis makes it possible to structure large data in ways that facilitate finding common patterns and mine-hidden information. In this study, we developed the population analysis with utilizing a correlation network model to conduct a study on stock market data on companies for the years 2000 through 2004. We utilized companies’ parameters for behavior assessment based on the population analysis. After creating the network model, we employed graph-based community algorithms, such as GLay, to identify communities and stocks with similar features associated with their excess returns. Our analysis of the top two communities revealed that companies in the finance sector have the highest share in the market, and companies with a low amount of capitalization h
ave a high excess return, similar to large companies. The proposed correlation network model and the associated population analysis show that investing in companies with high capitalization does not always guarantee higher rates of return on investment. Based on the proposed approach, investors could get similar returns by investing in certain small companies.
(More)