Export file:


  • RIS(for EndNote,Reference Manager,ProCite)
  • BibTex
  • Text


  • Citation Only
  • Citation and Abstract

An intrinsic robust rank-one-approximation approach for currency portfolio optimization

1 Department of Industrial Engineering, Tsinghua University, Beijing 100084, P.R. China
2 Department of Statistics, University of Wisconsin at Madison, Madison, WI 53706, USA

Special Issue: Financial Big Data Technology and Its Applications

A currency portfolio is a special kind of wealth whose value fluctuates with foreignexchange rates over time, which possesses 3Vs (volume, variety and velocity) properties of big datain the currency market. In this paper, an intrinsic robust rank one approximation (ROA) approachis proposed to maximize the value of currency portfolios over time. The main results of the paperinclude four parts: Firstly, under the assumptions about the currency market, the currency portfoliooptimization problem is formulated as the basic model, in which there are two types of variablesdescribing currency amounts in portfolios and the amount of each currency exchanged into another,respectively. Secondly, the rank one approximation problem and its variants are also formulated toapproximate a foreign exchange rate matrix, whose performance is measured by the Frobenius normor the 2-norm of a residual matrix. The intrinsic robustness of the rank one approximation is provedtogether with summarizing properties of the basic ROA problem and designing a modified powermethod to search for the virtual exchange rates hidden in a foreign exchange rate matrix. Thirdly,a technique for decision variables reduction is presented to attack the currency portfolio optimization.The reduced formulation is referred to as the ROA model, which keeps only variables describingcurrency amounts in portfolios. The optimal solution to the ROA model also induces a feasible solutionto the basic model of the currency portfolio problem by integrating forex operations from the ROAmodel with practical forex rates. Finally, numerical examples are presented to verify the feasibility ande ciency of the intrinsic robust rank one approximation approach. They also indicate that there existsan objective measure for evaluating and optimizing currency portfolios over time, which is related tothe virtual standard currency and independent of any real currency selected specially for measurement.
  Article Metrics


1.Bank for International Settlements, The Group of Ten. The Bank for International Settlements (BIS), 2016. Available from: https://www.bis.org/list/g10publications/.

2.Bapat RB, Raghavan TES (1997) Nonnegative Matrices and Applications. Cambridge University Press, Cambridge.

3.Brauer A (1957) A new proof of theorems of Perron and Frobenius on nonnegative matrix: I. positive matrices. DUKE MATH J 23: 367–378.

4.EI Ghaoui L, Lebret H (1997) Robust Solutions to Least Squares Problems with Uncertain Data. SIAM J MATRIX ANAL A 18: 1035–1064.    

5.Fabozzi FJ (2009) Institutional Investment Management: Equity and Bond Portfolio Strategies and Applications. John Wiley and Sons, Hoboken.

6.Gillis N, Franҫis G (2011) Low-Rank Matrix Approximation with Weights or Missing Data Is NPHard. SIAM J MATRIX ANAL A 32: 1149–1165.    

7.Gillis N, Franҫis G (2014) A Continuous Characterization of the Maximum-Edge Biclique Problem. J GLOBAL OPTIM 58: 439–464.    

8.Golub GH, Van Loan CF (1996) Matrix Computations. The Johns Hopkins University Press.

9.Housholder AS (1956) On The Convergence of Matrix Iterations. J ACM 3: 314–324.    

10.Huang H, Zhang Z (2016) Virtual Standard Currency for Approximating Foreign Exchange Rates. submitted to the International Journal of Electronic Commerce, accepted.

11.Housholder AS (1956) On The Convergence of Matrix Iterations. J ACM 3: 314–324.    

12.International Monetory Fund, Special Drawing Right SDR, 2016. Available from: http://www.imf.org/external/np/exr/facts/sdr.HTH.

13.Kannan R, Ishteva M, Drake B et al. (2016) Bounded Matrix Low Rank Approximation, In Ganesh R. Naik(ed.), Non-negative Matrix Factorization Techniques: Advances in Theory and Applications Springer Verlag.

14.Lee DD, Seung HS (2001) Algorithms for Non-negative Matrix Factorization, In T. K. Leen, T. G. Dietterich and V. Tresp, Advances in Neural Information Processing Systems 13, 556–562.

15.Markowitz HM (1952) Portfolio Selection. J Financ 7: 77–91.

16.Markowitz HM (1990) Normative Portfolio Analysis: Past, Present, and Future. J Econ Bus 42: 99–103.    

17.Naik GR (2016) Non-negative Matrix Factorization Techniques: advances in theory and applications, Springer Verlag.

18.Pennacchi G (2008) Theory of Asset Pricing, Pearson Addison Wesley, Baston.

19.Record N (2004) Currency Overlay, Wiley, New York.

20.Swiss Association for Standardization, Current currency and funds code list 4217-2015, 2016. Available from: http://www.currency-iso.org/en/home/tables/table-a1.html.

21.Takahashi N, Hibi R (2014) Global Convergence of Modified Multiplicative Updates for Nonnegative Matrix Factorization. Comput Optim Appl 57: 417–440.    

22.Vavasis SA (2009) On the complexity of nonnegative matrix factorization. SIAM J Optimiz 20: 1364–1377.

23.Yeandle M, MainelliM(2016) The Global Financial Centres Index 19. Z/Yen Group Limited. Available from: http://www.zyen.com/research/gfci.html.

24.Zeng L (2001) Some applications of spectral theory of nonnegative matrices to input-output models. Linear Algebra and Its Applications 336: 205–218.    

25.Zhang F (1999) Matrix Theory: Basic Results and Techniques, Spring-Verlag, New York.

© 2018 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution Licese (http://creativecommons.org/licenses/by/4.0)

Download full text in PDF

Export Citation

Article outline

Show full outline
Copyright © AIMS Press All Rights Reserved