Portfolio optimization with two quasiconvex risk measures

We study a static portfolio optimization problem with two risk measures: a principle risk measure in theobjective function and a secondary risk measure whose value is controlled in the constraints. This problem is of interestwhen it is necessary to consider the risk preferences of two parties, such as a portfolio manager and a regulator, at thesame time. A special case of this problem where the risk measures are assumed to be coherent (positively homogeneous)is studied recently in a joint work of the author. The present paper extends the analysis to a more general setting byassuming that the two risk measures are only quasiconvex. First, we study the case where the principal risk measure isconvex. We introduce a dual problem, show that there is zero duality gap between the portfolio optimization problem andthe dual problem, and finally identify a condition under which the Lagrange multiplier associated to the dual problemat optimality gives an optimal portfolio. Next, we study the general case without the convexity assumption and showthat an approximately optimal solution with prescribed optimality gap can be found by using the well-known bisectionalgorithm combined with a duality result that we prove.

PDF

___

[1] Agrawal A, Boyd SP. Disciplined quasiconvex programming. Optimization Letters (2020); 14: 1643-1657.
[2] Aktürk TD, Ararat Ç. Portfolio optimization with two coherent risk measures. Journal of Global Optimization 2020; 78 (3): 597-626.
[3] Artzner P, Delbaen F, Eber JM, Heath D. Coherent measures of risk. Mathematical Finance 1999; 9 (3): 203-228.
[4] Borwein JM, Lewis AS. Partially finite convex programming, Part I: Quasi relative interiors and duality theory. Mathematical Programming 1992; 57 (1): 15-48.
[5] Cerreia-Vioglio S, Maccheroni F, Marinacci M, Montrucchio L. Risk measures: rationality and diversification. Mathematical Finance 2011; 21 (4): 743-774.
[6] Drapeau S, Kupper M. Risk preferences and their robust representation. Mathematics of Operations Research 2013; 38 (1): 28-62.
[7] Föllmer H, Schied A. Stochastic Finance: An Introduction in Discrete Time. 4th revised edition. Berlin, Germany: De Gruyter, 2016.
[8] Grant M, Boyd SP. Graph implementations for nonsmooth convex programs. In: Blondel VD, Boyd SP, Kimura H (editors). Recent Advances in Learning and Control. London, UK: Springer, 2008, pp. 95-110.
[9] Källblad S. Risk- and ambiguity-averse portfolio optimization with quasiconvex utility functionals. Finance and Stochastics 2017; 21 (2): 397-425.
[10] Landsman Z. Minimization of the root of a quadratic functional under an affine equality constraint. Journal of Computational and Applied Mathematics 2008; 216: 319-327.
[11] Landsman Z, Makov U. Minimization of a function of a quadratic functional with application to optimal portfolio selection. Journal of Optimization Theory and Applications 2016; 170: 308-322.
[12] Mastrogiacomo E, Rosazza Gianin E. Portfolio optimization with quasiconvex risk measures, Mathematics of Operations Research 2015; 40 (4): 1042-1059.
[13] Owadally I. An improved closed-form solution for the constrained minimization of the root of a quadratic functional. Journal of Computational and Applied Mathematics 2011; 236: 4428-4435.
[14] Penot JP, Volle M. On quasiconvex duality. Mathematics of Operations Research 1990; 15 (4): 597-625.
[15] Rockafellar RT. Convex Analysis. Princeton, NJ, USA: Princeton University Press, 1970.
[16] Sion M. On general minimax theorems. Pacific Journal of Mathematics 1958; 8 (1): 171-176.
[17] Zălinescu C. Convex Analysis in General Vector Spaces. Singapore: World Scientific, 2002.