Abstract
This paper studies the mean-risk portfolio optimization problem with nonconvex transaction costs. We employ the conditional value-at-risk (CVaR) as a risk measure. There are a number of studies that aim at efficiently solving large-scale CVaR minimization problems. None of these studies, however, take into account nonconvex transaction costs, which are present in practical situations. To make a piecewise linear approximation of the transaction cost function, we utilized special ordered set type two constraints. Moreover, we devised a subgradient-based cutting plane algorithm to handle a large number of scenarios. This cutting plane algorithm needs to solve a mixed integer linear programming problem in each iteration, and this requires a substantial computation time. Thus, we also devised a two-phase cutting plane algorithm that is even more efficient. Numerical experiments demonstrated that our algorithms can attain near-optimal solutions to large-scale problems in a reasonable amount of time. Especially when rebalancing a current portfolio that is close to an optimal one, our algorithms considerably outperform other solution methods.
Similar content being viewed by others
Notes
This procedure aims at minimizing the right side of (12) with respect to \(\bar{a} \in \mathbb {R}\), and its effectiveness was confirmed through preliminary computational experiments.
References
Ágoston KC (2012) CVaR minimization by the SRA algorithm. Cent Eur J Oper Res 20:623–632
Ahmed S (2006) Convexity and decomposition of mean-risk stochastic programs. Math Program 106: 433–446
Alexander S, Coleman TF, Li Y (2006) Minimizing CVaR and VaR for a portfolio of derivatives. J Bank Financ 30:583–605
Angelelli E, Mansini E, Speranza MG (2008) A comparison of MAD and CVaR models with real features. J Bank Financ 32:1188–1197
Artzner P, Delbaen F, Eber JM, Heath D (1999) Coherent measures of risk. Math Financ 9:203–228
Beale EML, Tomlin JA (1970) Special facilities in a general mathematical programming system for non-convex problems using ordered sets of variables. In: Proceedings of the 5th international conference on operations research, pp 447–454
Beliakov G, Bagirov A (2006) Non-smooth optimization methods for computation of the conditional value-at-risk and portfolio optimization. Optimization 55:459–479
Bertsimas D, Darnell C, Soucy R (1999) Portfolio construction through mixed-integer programming at Grantham, Mayo, Van Otterloo and Company. Interfaces 29:49–66
Fábián CI (2008) Handling CVaR objectives and constraints in two-stage stochastic models. Eur J Oper Res 191:888–911
Fábián CI, Veszpémi A (2008) Algorithms for handling CVaR-constraints in dynamic stochastic programming models with applications to finance. J Risk 10:111–131
Iyengar G, Ma AKC (2013) Fast gradient descent method for mean-CVaR optimization. Ann Oper Res 205:203–212
Kaut M, Vladimirou H, Wallace SW, Zenios SA (2007) Stability analysis of portfolio management with conditional value-at-risk. Quant Financ 7:397–409
Kellerer H, Mansini R, Speranza MG (2000) Selecting portfolios with fixed costs and minimum transaction lots. Ann Oper Res 99:287–304
Konno H, Waki H, Yuuki A (2002) Portfolio optimization under lower partial risk measures. Asia Pac Financ Mark 9:127–140
Konno H, Yamamoto R (2005a) Global optimization versus integer programming in portfolio optimization under nonconvex transaction costs. J Global Optim 32:207–219
Konno H, Yamamoto R (2005b) Integer programming approaches in mean-risk models. Comput Manag Sci 2:339–351
Künzi-Bay A, Mayer J (2006) Computational aspects of minimizing conditional value-at-risk. Comput Manag Sci 3:3–27
Lim C, Sherali HD, Uryasev S (2008) Portfolio optimization by minimizing conditional value-at-risk via nondifferentiable optimization. Comput Optim Appl 46:391–415
Lobo MS, Fazel M, Boyd S (2007) Portfolio optimization with linear and fixed transaction costs. Ann Oper Res 152:341–365
Markowitz H (1952) Portfolio selection. J Financ 7:77–91
Perold AF (1984) Large-scale portfolio optimization. Manag Sci 30:1143–1160
Pflug G (2000) Some remarks on the value-at-risk and the conditional value-at-risk. In: Uryasev S (ed) Probabilistic constrained optimization: methodology and applications. Kluwer, Dordrecht, pp 272–281
Rockafellar RT, Uryasev S (2000) Optimization of conditional value-at-risk. J Risk 2:21–42
Rockafellar RT, Uryasev S (2002) Conditional value-at-risk for general loss distributions. J Bank Financ 26:1443–1471
Takeda A, Kanamori T (2009) A robust approach based on conditional value-at-risk measure to statistical learning problems. Eur J Oper Res 198:287–296
Tong X, Qi L, Wu F, Zhou H (2010) A smoothing method for solving portfolio optimization with CVaR and applications in allocation of generation asset. Appl Math Comput 216:1723–1740
Van Slyke RM, Wets R (1969) L-shaped linear programs with applications to optimal control and stochastic programming. SIAM J Appl Math 17:638–663
Yamamoto R, Konno H (2006) An efficient algorithm for solving a mean-variance model under nonconvex transaction costs. Pac J Optim 2:385–394
Yamamoto R, Konno H (2013) Rebalance schedule optimization of a large scale portfolio under transaction cost. J Oper Res Soc Japan 56:26–37
Zverovich V, Fábián CI, Ellison EF, Mitra G (2012) A computational study of a solver system for processing two-stage stochastic LPs with enhanced Benders decomposition. Math Program Comput 4:211–238
Acknowledgments
N. Sukegawa was supported by a Grant-in-Aid for Japan Society for the Promotion of Science (JSPS) Fellows.
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
We show that the complementary conditions, \(x^{+}_i x^{-}_i = 0,~\forall i = 1,2,\ldots ,I\), can be eliminated from problem (6).
Proposition 1
Suppose that \((\hat{a},\hat{u},\hat{{\varvec{x}}},\hat{{\varvec{x}}}^{+},\hat{{\varvec{x}}}^{-})\) is an optimal solution to problem (6) without the complementary conditions. Define \(\dot{x}^{+}_i := [\hat{x}_i - x^0_i]_{+}\) and \(\dot{x}^{-}_i := [x^0_i - \hat{x}_i]_{+}\) for \(i=1,2,\ldots ,I\). Then \((\hat{a},\hat{u},\hat{{\varvec{x}}},\dot{{\varvec{x}}}^{+},\dot{{\varvec{x}}}^{-})\) is an optimal solution to problem (6).
Proof
First, we show that \((\hat{a},\hat{u},\hat{{\varvec{x}}},\dot{{\varvec{x}}}^{+},\dot{{\varvec{x}}}^{-})\) is a feasible solution to problem (6). It follows from the definition that
In addition, since \(0 \le \hat{x}^{+}_i,~\hat{x}_i - x^0_i = \hat{x}^{+}_i - \hat{x}^{-}_i \le \hat{x}^{+}_i\) and \(0 \le \hat{x}^{-}_i,~x^0_i - \hat{x}_i = \hat{x}^{-}_i - \hat{x}^{+}_i \le \hat{x}^{-}_i\), we have
Considering that \(\mathcal{C}^{+}_i\) and \(\mathcal{C}^{-}_i\) are nondecreasing functions, we can see that
Thus, \((\hat{a},\hat{u},\hat{{\varvec{x}}},\dot{{\varvec{x}}}^{+},\dot{{\varvec{x}}}^{-})\) satisfies all the constraints of problem (6).
Now we show that \((\hat{a},\hat{u},\hat{{\varvec{x}}},\dot{{\varvec{x}}}^{+},\dot{{\varvec{x}}}^{-})\) is an optimal solution to problem (6). It follows from (17) that the objective function value of \((\hat{a},\hat{u},\hat{{\varvec{x}}},\dot{{\varvec{x}}}^{+},\dot{{\varvec{x}}}^{-})\) is not greater than that of \((\hat{a},\hat{u},\hat{{\varvec{x}}},\hat{{\varvec{x}}}^{+},\hat{{\varvec{x}}}^{-})\). Recall that \((\hat{a},\hat{u},\hat{{\varvec{x}}},\hat{{\varvec{x}}}^{+},\hat{{\varvec{x}}}^{-})\) is an optimal solution to a relaxed problem, i.e., problem (6) without complementary conditions. This completes the proof. \(\square \)
Remark 1
When \(\mathcal{C}^{+}_i\) and \(\mathcal{C}^{-}_i\) are strictly increasing functions, \((\hat{a},\hat{u},\hat{{\varvec{x}}},\hat{{\varvec{x}}}^{+},\hat{{\varvec{x}}}^{-})\) satisfies the complementary conditions by itself. This is because if \(i\) exists such that \(\hat{x}^{+}_i \hat{x}^{-}_i \not = 0\), the objective function value of \((\hat{a},\hat{u},\hat{{\varvec{x}}},\dot{{\varvec{x}}}^{+},\dot{{\varvec{x}}}^{-})\) is smaller than that of \((\hat{a},\hat{u},\hat{{\varvec{x}}},\hat{{\varvec{x}}}^{+},\hat{{\varvec{x}}}^{-})\), which contradicts the optimality of \((\hat{a},\hat{u},\hat{{\varvec{x}}},\hat{{\varvec{x}}}^{+},\hat{{\varvec{x}}}^{-})\).
Rights and permissions
About this article
Cite this article
Takano, Y., Nanjo, K., Sukegawa, N. et al. Cutting plane algorithms for mean-CVaR portfolio optimization with nonconvex transaction costs. Comput Manag Sci 12, 319–340 (2015). https://doi.org/10.1007/s10287-014-0209-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10287-014-0209-7