Skip to main content
Log in

Aumann–Serrano index of risk in portfolio optimization

  • Original Article
  • Published:
Mathematical Methods of Operations Research Aims and scope Submit manuscript

Abstract

The paper is devoted to study the portfolio optimization problem for an investor who aims to minimize the exposure to equity markets measured by the Aumann–Serrano index of riskiness. The ARMA–GARCH model with normal variance–mean mixture innovations is employed to capture the stylized facts of stock returns. Using a two-step scheme, we convert the high-dimensional optimization problem into a two-dimensional one. We further prove that the dimension reduction technique preserves the convexity of the problem as long as the risk measure is convex and monotonic. In the empirical study, we observe that the optimal portfolio outperforms benchmarks based on a 10-year backtesting window covering the financial crisis.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2

Similar content being viewed by others

Notes

  1. It is more precise to say GH and NTS distributions are semi-heavy-tailed because both have exponentially decaying tails.

  2. The lower bound of \(\mu \) is derived by setting the expectation, or the derivative of MGF (13) at \(t=0\), to be zero. The upper bound is derived by setting \(M_X(l)\ge 1\) in (15).

  3. Typically, the first step of QMLE estimation of ARMA–GARCH parameters is performed assuming normal innovations. The QMLE with Student’s t innovations is proved to be practically useful in Kim et al. (2010), Anand et al. (2016) and Goode et al. (2015).

  4. Based on the definition formula (1), we have \(cov({\varvec{X}}) = \mathbb {E}(T){\varvec{\varSigma }} + var(T){\varvec{\beta }}{\varvec{\beta }}'\).

  5. The results in this empirical analysis are mainly for illustration purpose. To address practitioners’ concerns, one should incorporate transaction cost in the optimization. As we rebalance the portfolio on a daily basis, the impact of transaction cost is not negligible.

References

  • Abramowitz M, Stegun IA (1964) Handbook of mathematical functions: with formulas, graphs, and mathematical tables, vol 55. Courier Corporation, North Chelmsford

    MATH  Google Scholar 

  • Acerbi C, Tasche D (2002) Expected shortfall: a natural coherent alternative to value at risk. Econ Notes 31:379–388

    Article  Google Scholar 

  • Anand A, Li T, Kurosaki T, Kim YS (2016) Foster–Hart optimal portfolios. J Bank Finance 68:117–130

    Article  Google Scholar 

  • Artzner P, Delbaen F, Eber JM, Heath D (1999) Coherent measures of risk. Math Finance 9(3):203–228

    Article  MathSciNet  Google Scholar 

  • Aumann RJ, Serrano R (2008) An economic index of riskiness. J Polit Econ 116(5):810–836

    Article  Google Scholar 

  • Barndorff-Nielsen O (1977) Exponentially decreasing distributions for the logarithm of particle size. Proc R Soc Lond 353(1674):401–409

    Google Scholar 

  • Barndorff-Nielsen O, Kent J, Sorensen M (1982) Normal variance–mean mixtures and z distributions. Int Stat Rev 50(2):145–159

    Article  MathSciNet  Google Scholar 

  • Barndorff-Nielsen O, Levendorskii S et al (2001) Feller processes of normal inverse Gaussian type. Quant Finance 1(3):318–331

    Article  MathSciNet  Google Scholar 

  • Berkowitz J, O’Brien J (2002) How accurate are value-at-risk models at commercial banks? J Finance 57(3):1093–1111

  • Cont R (2001) Empirical properties of asset returns: stylized facts and statistical issues. Quant Finance 1(2):223–236

    Article  Google Scholar 

  • Cvitanic J, Karatzas I (1995) On portfolio optimization under “drawdown” constraints. IMA Vol Math Appl 65:35

  • Fama EF (1963) Mandelbrot and the stable Paretian hypothesis. J Bus 36(4):420–429

    Article  Google Scholar 

  • Foster DP, Hart S (2009) An operational measure of riskiness. J Polit Econ 117(5):785–814

    Article  Google Scholar 

  • Goode J, Kim YS, Fabozzi FJ (2015) Full versus quasi MLE for ARMA–GARCH models with infinitely divisible innovations. Appl Econ Inc Appl Financ Econ 47(48):5147–5158

    Google Scholar 

  • Grossman SJ, Zhou Z (1993) Optimal investment strategies for controlling drawdowns. Math Finance 3(3):241–276

    Article  Google Scholar 

  • Homm U, Pigorsch C (2012a) Beyond the Sharpe ratio: an application of the Aumann–Serrano index to performance measurement. J Bank Finance 36(8):2274–2284

    Article  Google Scholar 

  • Homm U, Pigorsch C (2012b) An operational interpretation and existence of the Aumann–Serrano index of riskiness. Econ Lett 114(3):265–267

    Article  MathSciNet  Google Scholar 

  • Jorion P (1997) Value at risk: the new benchmark for controlling market risk. Irwin Professional Publishing, Chicago

    Google Scholar 

  • Kim YS, Rachev ST, Bianchi ML, Fabozzi FJ (2010) Tempered stable and tempered infinitely divisible GARCH models. J Bank Finance 34(9):2096–2109

    Article  Google Scholar 

  • Kim YS, Rachev ST, Bianchi ML, Mitov I, Fabozzi FJ (2011) Time series analysis for financial market meltdowns. J Bank Finance 35(8):1879–1891

    Article  Google Scholar 

  • Kim YS, Lee J, Mittnik S, Park J (2015) Quanto option pricing in the presence of fat tails and asymmetric dependence. J Econom 187(2):512–520

    Article  MathSciNet  Google Scholar 

  • Mandelbrot B (1963) New methods in statistical economics. J Polit Econ 71(5):421–440

    Article  Google Scholar 

  • Markowitz H (1952) Portfolio selection. J Finance 7(1):77–91

    Google Scholar 

  • Rachev ST, Kim YS, Bianchi ML, Fabozzi FJ (2011) Financial models with Lévy processes and volatility clustering, vol 187. Wiley, New York

    Book  Google Scholar 

  • Rockafellar RT, Uryasev S (2000) Optimization of conditional value-at-risk. J Risk 2:21–42

    Article  Google Scholar 

  • Schulze K (2014) Existence and computation of the Aumann–Serrano index of riskiness and its extension. J Math Econ 50:219–224

    Article  MathSciNet  Google Scholar 

  • Sharpe WF (1994) The Sharpe ratio. J Portf Manag 21(1):49–58

    Article  Google Scholar 

  • Shephard N, Barndorff-Nielsen OE (2001) Normal modified stable processes. Technical report, University of Oxford, Department of Economics

  • Shi X, Kim YS (2021) Coherent risk measure and normal mixture distributions with application in portfolio optimization. Int J Theor Appl Finance 24 (4): 2150019

Download references

Acknowledgements

We thank Abhinav Anand, Tetsuo Kurosaki and Xiang Shi for valuable discussions over the course of writing this paper. We thank two anonymous reviewers whose comments and suggestions helped improve and clarify the manuscript. The work was supported by National Natural Science Foundation of China (Grant No. 72071132, 72173089, 71773079) and Shenzhen University (Grant No. SZJR003).

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Fumin Zhu.

Ethics declarations

Conflict of interest

The opinions and statements expressed in this paper are those of the authors and may be different to views or opinions otherwise held by Barclays.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendices

Appendices

A Proof of Proposition 1

Proof

To solve problem (28), we need to prove the following first

$$\begin{aligned} \underset{{\varvec{w}}}{\text {argmin }} R({\tilde{\mu }}+{\tilde{\beta }}T+\sqrt{T}\sigma ^{(p)}Z) = \underset{{\varvec{w}}}{\text {argmin }}\sigma ^{(p)}. \end{aligned}$$
(42)

Let \(R(\sigma ):=R({\tilde{\mu }} + {\tilde{\beta }} T + \sigma \sqrt{T}Z)\). Notice that \(\sigma Z\) and \(-\sigma Z\) have exactly the same distribution, where Z is a univariate standard Gaussian random variable. The domain of \(R(\sigma )\) can be extended from \(\mathbb {R}^+\) to \(\mathbb {R}\), and it is trivially true that \(R(\sigma )\) is symmetric to \(\sigma =0\).

Since \(R(\cdot )\) is a convex risk measure, for any \(0< \lambda <1\) and \(\sigma _1, \sigma _2\in \mathbb {R}\) we have

$$\begin{aligned}&R\left( \lambda \sigma _1 + (1-\lambda )\sigma _2\right) \\&\quad = R\left( {\tilde{\mu }} + {\tilde{\beta }} T + [\lambda \sigma _1 + (1-\lambda )\sigma _2]\sqrt{T}Z\right) \\&\quad = R\left( \lambda [{\tilde{\mu }} + {\tilde{\beta }} T + \sigma _1\sqrt{T}Z] + (1-\lambda )[{\tilde{\mu }} + {\tilde{\beta }} T + \sigma _2\sqrt{T}Z] \right) \\&\quad \le \lambda R({\tilde{\mu }} + {\tilde{\beta }} T + \sigma _1\sqrt{T}Z) + (1-\lambda )R({\tilde{\mu }} + {\tilde{\beta }} T + \sigma _2\sqrt{T}Z) \\&\quad = \lambda R(\sigma _1) + (1-\lambda )R(\sigma _2). \end{aligned}$$

Therefore, \(R(\sigma )\) is a convex function of \(\sigma \). If there exist \(0<\sigma _1<\sigma _2\), and \(R(\sigma _1)>R(\sigma _2)\), then find \(\lambda \) such that \(\lambda (-\sigma _2) + (1-\lambda )\sigma _2 = \sigma _1\). However,

$$\begin{aligned} \lambda R(-\sigma _2) + (1-\lambda ) R(\sigma _2) = R(\sigma _2) < R(\sigma _1), \end{aligned}$$

which contradicts with the convexity of \(R(\sigma )\). In other words, \(R(\sigma )\) is nondecreasing on \(\mathbb {R}^+\), so (42) holds.

Recall that \(\sigma ^{(p)}:=\sqrt{{\varvec{w}}'{\varvec{\varSigma }} {\varvec{w}}}\), and also by (42), problem (28) is therefore equivalent to the linear-quadratic programming below

$$\begin{aligned} \begin{aligned}&\underset{{\varvec{w}}}{\text {min}}&\frac{1}{2}{\varvec{w}}'{\varvec{\varSigma }} {\varvec{w}} \\&\text {s.t.}&{\varvec{w}}'{\varvec{e}} = 1, \\&{\varvec{w}}'{\varvec{\mu }} = {\tilde{\mu }}, \\&{\varvec{w}}'{\varvec{\beta }} = {\tilde{\beta }}. \end{aligned} \end{aligned}$$
(43)

A simple application of Lagrange multiplier will prove the proposition. \(\square \)

B Proof of Proposition 2

To prove Proposition 2, we need to prove the following lemma first.

Lemma 1

If \({\varvec{A}}\) is positive definite, then \(f({\varvec{x}}) = \sqrt{{\varvec{x}}'{\varvec{A}}{\varvec{x}}}\) is convex on \(\mathbb {R}^d\).

Proof

Define an inner product space \(\langle \cdot ,\cdot \rangle : \mathbb {R}^{d} \times \mathbb {R}^d \rightarrow \mathbb {R}\) such that \(\langle {\varvec{x}},{\varvec{y}} \rangle = {\varvec{x}}'{\varvec{A}}{\varvec{y}}\), for any \( {\varvec{x}}, {\varvec{y}} \in \mathbb {R}^d\). This inner product space is well-defined since for any \({\varvec{x}},{\varvec{y}},{\varvec{z}} \in \mathbb {R}^d\),

  1. 1.

    \(\langle {\varvec{x}},{\varvec{y}} \rangle = {\varvec{x}}'{\varvec{A}}{\varvec{y}}=({\varvec{x}}'{\varvec{A}}{\varvec{y}})'={\varvec{y}}'{\varvec{A}}{\varvec{x}}=\langle {\varvec{y}},{\varvec{x}} \rangle \)

  2. 2.

    \(\langle a{\varvec{x}},{\varvec{y}} \rangle =a{\varvec{x}}'{\varvec{A}}{\varvec{y}}=a\langle {\varvec{x}},{\varvec{y}} \rangle \), \(\langle {\varvec{x}}+{\varvec{y}},{\varvec{z}} \rangle =({\varvec{x}}+{\varvec{y}})'{\varvec{A}}{\varvec{z}}={\varvec{x}}'{\varvec{A}}{\varvec{z}}+{\varvec{y}}'{\varvec{A}}{\varvec{z}}=\langle {\varvec{x}},{\varvec{z}} \rangle +\langle {\varvec{y}},{\varvec{z}} \rangle \)

  3. 3.

    \(\langle {\varvec{x}},{\varvec{x}} \rangle = {\varvec{x}}'{\varvec{A}}{\varvec{x}}>0, \ \forall \ {\varvec{x}} \in \mathbb {R}^d \setminus \{{\varvec{0}}\}\)

A simple inequality follows from the Cauchy–Schwarz inequality:

$$\begin{aligned} \sqrt{\langle \lambda {\varvec{x}} + (1-\lambda ){\varvec{y}}, \lambda {\varvec{x}} + (1-\lambda ){\varvec{y}} \rangle } \le \lambda \sqrt{\langle {\varvec{x}}, {\varvec{x}} \rangle } + (1-\lambda )\sqrt{\langle {\varvec{y}}, {\varvec{y}} \rangle }, \ \forall \lambda \in [0,1], \end{aligned}$$

which proves the convexity. \(\square \)

Now we can prove Proposition 2.

Proof

Let \(h({\tilde{\mu }},{\tilde{\beta }}, {\tilde{\gamma }}) = ({\tilde{\mu }}\ {\tilde{\beta }}\ {\tilde{\gamma }}){\varvec{\varOmega }}^{-1}({\tilde{\mu }}\ {\tilde{\beta }}\ {\tilde{\gamma }})'\) where \({\varvec{\varOmega }} = ({\varvec{\mu }}\ {\varvec{\beta }}\ {\varvec{e}})'{\varvec{\varSigma }}^{-1}({\varvec{\mu }}\ {\varvec{\beta }}\ {\varvec{e}})\), and \({\varvec{\varSigma }}\) and \({\varvec{\varSigma }}^{-1}\) are positive definite. By Lemma 1, for any \(0\le \lambda \le 1\),

$$\begin{aligned}&\sqrt{h\left( \lambda {\tilde{\mu }}_1 + (1-\lambda ) {\tilde{\mu }}_2,\lambda {\tilde{\beta }}_1 + (1-\lambda ) {\tilde{\beta }}_2,\lambda {\tilde{\gamma }}_1 + (1-\lambda ) {\tilde{\gamma }}_2 \right) } \\&\quad \le \ \lambda \sqrt{h({\tilde{\mu }}_1,{\tilde{\beta }}_1, {\tilde{\gamma }}_1 )} + (1-\lambda )\sqrt{h({\tilde{\mu }}_2,{\tilde{\beta }}_2, {\tilde{\gamma }}_2 )}. \end{aligned}$$

Since \(h({\tilde{\mu }},{\tilde{\beta }}, 1) = g({\tilde{\mu }},{\tilde{\beta }})\), we obtain that

$$\begin{aligned}&\sqrt{g\left( \lambda {\tilde{\mu }}_1 + (1-\lambda ) {\tilde{\mu }}_2,\lambda {\tilde{\beta }}_1 + (1-\lambda ) {\tilde{\beta }}_2 \right) } \\&\quad \le \ \lambda \sqrt{g({\tilde{\mu }}_1,{\tilde{\beta }}_1 )} + (1-\lambda )\sqrt{g({\tilde{\mu }}_2,{\tilde{\beta }}_2 )}. \end{aligned}$$

Therefore, \(\sqrt{g({\tilde{\mu }},{\tilde{\beta }})}\) is convex on \(\mathbb {R}^2\).

We denote \(R({\tilde{\mu }},{\tilde{\beta }}) = R\left( {\tilde{\mu }}+{\tilde{\beta }}T+\sqrt{g({\tilde{\mu }},{\tilde{\beta }})}\sqrt{T}Z\right) \), for any \(0\le \lambda \le 1\),

$$\begin{aligned}&\ \lambda R({\tilde{\mu }}_1,{\tilde{\beta }}_1)+(1-\lambda )R({\tilde{\mu }}_2,{\tilde{\beta }}_2) \nonumber \\&\quad = \ \lambda R\left( {\tilde{\mu }}_1+{\tilde{\beta }}_1T+\sqrt{g({\tilde{\mu }}_1,{\tilde{\beta }}_1)}\sqrt{T}Z\right) +(1-\lambda ) R\left( {\tilde{\mu }}_2+{\tilde{\beta }}_2T+\sqrt{g({\tilde{\mu }}_2,{\tilde{\beta }}_2)}\sqrt{T}Z\right) \nonumber \\&\quad \ge \ R\left( {\bar{\mu }}+{\bar{\beta }}T+\left[ \lambda \sqrt{g({\tilde{\mu }}_1,{\tilde{\beta }}_1)}+(1-\lambda )\sqrt{g({\tilde{\mu }}_2,{\tilde{\beta }}_2)}\right] \sqrt{T}Z\right) , \end{aligned}$$
(44)

where \({\bar{\mu }} = \lambda {\tilde{\mu }}_1+(1-\lambda ) {\tilde{\mu }}_2\) and \({\bar{\beta }} = \lambda {\tilde{\beta }}_1+(1-\lambda ) {\tilde{\beta }}_2\), and the last inequality holds by the convexity of \(R(\cdot )\).

If we further denote \(\alpha =\lambda \sqrt{g({\tilde{\mu }}_1,{\tilde{\beta }}_1)}+(1-\lambda )\sqrt{g({\tilde{\mu }}_2,{\tilde{\beta }}_2)}\) and \(\gamma = \sqrt{g\left( {\bar{\mu }},{\bar{\beta }}\right) }\), then \(\alpha \ge \gamma \ge 0\) by the convexity of \(\sqrt{g(\cdot , \cdot )}\). Consider two normal variance–mean mixtures, \(X_{\alpha } = {\bar{\mu }}+{\bar{\beta }}T+\alpha \sqrt{T}Z\) and \(X_{\gamma } = {\bar{\mu }}+{\bar{\beta }}T+\gamma \sqrt{T}Z\). \(F_{X_{\alpha }}\) and \(F_{X_{\gamma }}\) are CDFs of \(X_{\alpha }\) and \(X_{\gamma }\) respectively. For any \(l \in \mathbb {R}\), we have

$$\begin{aligned}&\int _{-\infty }^l \left( F_{X_{\alpha }}(k)-F_{X_{\gamma }}(k)\right) \mathrm {d}k \\&\quad = \int _{-\infty }^l \left( \mathbb {P}({\bar{\mu }}+{\bar{\beta }}T+\alpha \sqrt{T}Z \le k)-\mathbb {P}({\bar{\mu }}+{\bar{\beta }}T+\gamma \sqrt{T}Z \le k) \right) \mathrm {d}k \\&\quad = \int _{-\infty }^l \mathbb {E}\left[ \mathbb {P}({\bar{\mu }}+{\bar{\beta }}T+\alpha \sqrt{T}Z \le k)-\mathbb {P}({\bar{\mu }}+{\bar{\beta }}T+\gamma \sqrt{T}Z \le k) | T\right] \mathrm {d}k\\&\quad = \int _{-\infty }^l \left( \int _0^{\infty } \left[ \mathbb {P}({\bar{\mu }}+{\bar{\beta }}T+\alpha \sqrt{t}Z \le k)-\mathbb {P}({\bar{\mu }}+{\bar{\beta }}T+\gamma \sqrt{t}Z \le k)\right] f_T(t)\mathrm {d}t \right) \mathrm {d}k\\&\quad = \int _0^{\infty } \left( \int _{-\infty }^l \left[ \mathbb {P}({\bar{\mu }}+{\bar{\beta }}t+\alpha \sqrt{t}Z \le k)-\mathbb {P}({\bar{\mu }}+{\bar{\beta }}t+\gamma \sqrt{t}Z \le k)\right] \mathrm {d}k \right) f_T(t)\mathrm {d}t\\&\quad = \int _0^{\infty } \left( \int _{-\infty }^l \left[ \mathbb {P} \left( \alpha Z \le \frac{k-{\bar{\mu }}-{\bar{\beta }}t}{\sqrt{t}}\right) -\mathbb {P} \left( \gamma Z \le \frac{k-{\bar{\mu }}-{\bar{\beta }}t}{\sqrt{t}} \right) \right] \mathrm {d}k \right) f_T(t)\mathrm {d}t\\&\quad = \int _0^{\infty } \left( \int _{-\infty }^{\frac{l-{\bar{\mu }}-{\bar{\beta }}t}{\sqrt{t}}} \left[ P \left( \alpha Z \le m \right) -P \left( \gamma Z \le m \right) \right] \mathrm {d}m \right) \sqrt{t} f_T(t)\mathrm {d}t\\&\quad \ge 0, \end{aligned}$$

where the interchangeability of integrals is guaranteed by Fubini’s Theorem. By definition, this means \(X_{\gamma }\) second-order stochastically dominates \(X_{\alpha }\). Therefore \(R(X_{\gamma }) \le R(X_{\alpha })\) by the monotonicity of \(R(\cdot )\), which in combination with (44) proves the convexity of \(R({\tilde{\mu }},{\tilde{\beta }})\) on \(\mathbb {R}^2\). \(\square \)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Li, T., Kim, Y.S., Fan, Q. et al. Aumann–Serrano index of risk in portfolio optimization. Math Meth Oper Res 94, 197–217 (2021). https://doi.org/10.1007/s00186-021-00753-x

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00186-021-00753-x

Keywords

Mathematics Subject Classification

Navigation