Robust Growth-Optimal Portfolios

Size: px
Start display at page:

Download "Robust Growth-Optimal Portfolios"

Transcription

1 Robust Growth-Optimal Portfolios! Daniel Kuhn! Chair of Risk Analytics and Optimization École Polytechnique Fédérale de Lausanne rao.epfl.ch

2 4 Technology Stocks I 4 technology companies: Intel, Cisco, Blackberry, and Nokia. I Monthly statistics: µ [%] INTC 0.37 CSCO 0.33 BBRY 0.93 NOK 0.85 [%] INTC CSCO BBRY NOK INTC CSCO BBRY NOK

3 Markowitz Portfolio Theory

4 Fixed-Mix Strategy w t (r,...,r t )=w I Keep portfolio weights constant over time. I Memoryless but dynamic.

5 Fixed-Mix Strategy in BS Economy I What to expect:

6 Fixed-Mix Strategy in BS Economy I What really happens:

7 Fixed-Mix Strategy in BS Economy I What really happens:

8 Fixed-Mix Strategy in BS Economy I What really happens:

9 All is Lost w.p.!

10 Asymptotic Winners & Losers Portfolio growth rate = expected log utility (w) =E log ( + w r) CLT: Growth rate determines porfolio performance in the long run. I I (w) > 0: asymptotic winners (w) < 0: asymptotic losers Growth optimal portfolio (GOP) is the one that maximizes (w). w g 2 argmax w (w) Kelly & Latané: GOP outperforms any other non-anticipative investment strategy with probability in the long run. Kelly, Bell Syst. Tech. J Latané, JPE 959

11 Markowitz Winners & Losers

12 Distributional Assumptions I "µ and are the only quantities that can be distilled out of the past." I "The slightest acquaintance with problems of analyzing economic time series will suggest that this assumption is optimistic rather than unnecessarily restrictive." Roy (952) Roy, Econometrica 952

13 The long run may be indeed long... In a Black Scholes economy, it may take the GOP I 208 years to beat an all-cash strategy I 4,700 years to beat an all-stock strategy with 95% confidence. Rubinstein (99) Rubinstein, JPM 99

14 Robust Growth-Optimal Portfolios

15 Finite Investment Horizons Informally, our goal is to maximize terminal wealth TY t= ( + w r t ), which is equivalent to maximizing the growth rate T TX log ( + w r t ). t=

16 Finite Investment Horizons Formally, we can maximize the VaR of the growth rate VaR (w) = max ( : P T TX log ( + w r t )! ). t=

17 Distributional Robustness Stochastic and distributionally robust performance measures: I VaR (w) = max ( : P T TX log ( + w r t )! ). t= I WVaR (w) = max ( : P T TX log ( + w r t )! 8P 2 P ). t=

18 Robust GOP Performance Guarantees The fixed-mix strategy w will grow at least by et WVaR (w ) w.p. under any distribution P 2 P. The robust GOP (RGOP) is the portfolio that maximizes WVaR (w ). w 2 argmax WVaR (w ) w

19 Calculating WVaR (w) I Weak sense white noise ambiguity set I WVaR (w) P = P : E P ( r t )=µ, E P r s r t = st + µµ max ( : P T TX t= w r t 2 (w r! t ) 2 8P 2 P ) I Distributionally robust quadratic chance constraint.

20 Chance Constraints

21 Robust Individual Chance Constraints P(L(x, ) apple 0) 8P 2 P inf P2P m P(L(x, ) apple 0)

22 ASimpleExample Individual chance constraint: P(x ξ + x 2 ξ 2 ) 75% ( ).5 Model : ξ i N (0, ) iid ( ) issocprepresentable: ( ) x 2 /Φ (75%) x x E(ξ )=E(ξ 2 )=0, Var(ξ )=Var(ξ 2 )=, Cov(ξ, ξ 2 )=0 ( )

23 ASimpleExample Individual chance constraint: P(x ξ + x 2 ξ 2 ) 75% ( ).5 Model 2: ξ i 2 δ + 2 δ + iid ( ) x x 2 x + x 2 feasible set is unbounded nonconvex x x E(ξ )=E(ξ 2 )=0, Var(ξ )=Var(ξ 2 )=, Cov(ξ, ξ 2 )=0 ( )

24 ASimpleExample Individual chance constraint: P(x ξ + x 2 ξ 2 ) 75% ( ) Model 3: ξ i κ2 +κ 2 δ κ feasible set changes discontinuously with κ + +κ 2 δ +κ iid x x E(ξ )=E(ξ 2 )=0, Var(ξ )=Var(ξ 2 )=, Cov(ξ, ξ 2 )=0 ( )

25 ASimpleExample Individual chance constraint: P(x ξ + x 2 ξ 2 ) 75% ( ) Model 3: ξ i κ2 +κ 2 δ κ feasible set changes discontinuously with κ + +κ 2 δ +κ iid x x E(ξ )=E(ξ 2 )=0, Var(ξ )=Var(ξ 2 )=, Cov(ξ, ξ 2 )=0 ( )

26 ASimpleExample Individual chance constraint: P(x ξ + x 2 ξ 2 ) 75% ( ).5 Model n: ξ, ξ follow any distribution satisfying ( ) 0.5 x x E(ξ )=E(ξ 2 )=0, Var(ξ )=Var(ξ 2 )=, Cov(ξ, ξ 2 )=0 ( )

27 ASimpleExample Individual chance constraint: P(x ξ + x 2 ξ 2 ) 75% ( ).5 Model n: ξ, ξ follow any distribution satisfying ( ) 0.5 x x E(ξ )=E(ξ 2 )=0, Var(ξ )=Var(ξ 2 )=, Cov(ξ, ξ 2 )=0 ( )

28 ASimpleExample Individual chance constraint: P(x ξ + x 2 ξ 2 ) 75% ( ).5 Model n: ξ, ξ follow any distribution satisfying ( ) 0.5 x x E(ξ )=E(ξ 2 )=0, Var(ξ )=Var(ξ 2 )=, Cov(ξ, ξ 2 )=0 ( )

29 ASimpleExample Individual chance constraint: inf P(x ξ + x 2 ξ 2 ) 75% P P ( ).5 Let P be the set of all distributions satisfying ( ) intersection of feasible sets is convex x x E(ξ )=E(ξ 2 )=0, Var(ξ )=Var(ξ 2 )=, Cov(ξ, ξ 2 )=0 ( )

30 Distributional Robustness Classical chance constraint: P(L(x, ξ) 0) ϵ hedges only against risk sensitive to estimation errors often nonconvex Distributionally robust chance constraint: inf P P P(L(x, ξ) 0) ϵ hedges against risk and ambiguity often convex Calafiore & El Ghaoui, JOTA 2006

31 CVaR Approximation 2 CVaR approximation: P-CVaR ϵ (L(x, ξ)) 0 P-VaR ϵ (L(x, ξ)) 0 P(L(x, ξ) 0) ϵ Thus, the CVaR constraint ϵ VaR ϵ CVaR ϵ 0 Loss P-CVaR ϵ (L(x, ξ)) 0 conservatively approximates the chance constraint P(L(x, ξ) 0) ϵ 2 Nemirovski & Shapiro, SIOPT 2006

32 Exactness of the CVaR Approximation 3 P = {P : E P (ξ) =µ, E P (ξξ T )=Σ + µµ T} Theorem If L(x, ξ) is either concave or quadratic in ξ, then sup P P inf P P P-CVaR ϵ (L(x, ξ)) 0 P (L(x, ξ) 0) ϵ Corollary If L(x, ξ) is either concave or quadratic in ξ, then sup P P = sup P P P-CVaR ϵ (L(x, ξ)) P-VaR ϵ (L(x, ξ)) 3 Zymler, K & Rustem, MPA 202

33 Tractability of the CVaR Approximation P = {P : E P (ξ) =µ, E P (ξξ T )=Σ + µµ T} Theorem If L(ξ) = sup P P min {a m + b m ξ}, then m=,...,l P-CVaR ϵ (L(ξ)) = inf β + ϵ Ω, M s.t. β R, M S k+ +, λ R l + l [ 0 M λ b ] 2 m m 0 2 bt m a m β m= l λ m = m=

34 Tractability of the CVaR Approximation P = {P : E P (ξ) =µ, E P (ξξ T )=Σ + µµ T} Theorem If L(ξ) =ξ Qξ + q ξ + q 0,then sup P P P-CVaR ϵ (L(ξ)) = inf β + ϵ Ω, M s.t. β R, M S k+ + [ Q M q ] 2 2 qt q 0 0 β

35 Robust Growth-Optimal Portfolios

36 Calculating WVaR (w) I Weak sense white noise ambiguity set I WVaR (w) P = P : E P ( r t )=µ, E P r s r t = st + µµ max ( : P T TX t= w r t 2 (w r! t ) 2 8P 2 P ) I Distributionally robust quadratic chance constraint.

37 Calculating WVaR (w) Re-express the distributionally robust chance constraint as an LMI. WVaR (w) = max s. t. M 2 S nt +, 2 R, 2 R + " 2 M h, Mi apple0, M 0 P T P t= P t ww T P t 2 t= P t w 2 PT t= P t w T # I Tractable, but... I The dimension of the LMIs is (nt + ) (nt + ). I n = 30, T = 20! 6.5M. decision variables. is the second-order moment matrix of [ r, r 2,..., r T ]. 2 P t 2 R n nt is a truncation operator: P t [r,...,r T ] = r t.

38 Calculating WVaR (w) Simplify the SDP using a projection property of distribution families. r t 2 R n (µ, ) t 2 R (w µ, w w) WVaR (w) = max s. t. M 2 S T +, 2 R, 2 R + M I More tractable. h (w), Mi apple0, M 0 apple 2 I 2 2 T I The dimension of the LMIs is (T + ) (T + ). I n = 30, T = 20! 7.4K decision variables. Yu et al., NIPS 2009

39 Calculating WVaR (w) Compound symmetric solution Initial M Compound symmetric M I Highly tractable. I The number of decision variables (M,, ) reduces to 6!

40 Calculating WVaR (w) Positive semidefiniteness of any compound symmetric matrix M with parameters, 2, 3, and 4 can be verified very efficiently. M 0 () 8 2 >< 4 0 +(T ) 2 0 >: 4 ( +(T ) 2 ) T 3 2

41 Calculating WVaR (w) WVaR (w) = max s. t. 2 R 4, 2 R, 2 R h i T 2p + µ p 2 + T (T )µ 2 p 2 + 2T µ p apple (T ) ( +(T ) 2 ) T T (T ) 2 0 ( 4 T + ) 2 +(T ) 2 T I Nonlinear program with 6 variables and 9 constraints!

42 Calculating WVaR (w) I Analytical solution: WVaR (w) = 2 w µ + r T /2 w! 2 T T w wa I Maximizing WVaR (w) gives rise to an SOCP whose size is independent of T.

43 Discussion I Relation to Markowitz RGOP is Markowitz-efficient, tailored to T and. I Long-term investors When T!, WVaR (w) reduces to 2 2 ( w µ) 2 {z } nominal growth rate 2 w w {z } risk premium I Relation to El Ghaoui et al. When T =, WVaR (w) reduces to w µ + r /2 w {z } formula by El Ghaoui 2 C A C A El Ghaoui et al., OR 2003

44 More Information: Support I Weak sense white noise ambiguity set with support I E.g. ellipsoidal support P = P \ P : P r,...,r T 2 = = ( r,...,r T : T TX t= (r t ) (r t ) apple )

45 More Information: Support WVaR (w) = max s. t. A 2 S n, B 2 S n, c 2 R n, d 2 R, 0, apple 0, 0, 2 R + (T ha, + µµ i + T (T ) hb, µµ i + 2T c µ + d) apple 0 apple apple A +(T )B c c d T A B 2 A +(T )B + c w 4 c + d+ ( ) 2 T w 2 apple A B + w w I The size of this SDP is independent of T.

46 Less Information: Moment Uncertainty I Moment estimates can differ greatly from the true values. I 2 nd -layer of robustness: confidence region. n (µ, ) :(µ ˆµ) ˆ (µ ˆµ) apple, ˆ o 2 I Analytical expression for WVaR (w) w ˆµ + p r! ( ) 2 + T ˆ /2 w! 2 2 (T ) w ˆ w T

47 Synthetic Experiment: Horizon Effects I Black Scholes economy I Calculate µ and from 0 Industry Portfolios [] I Compare VaRs and Sharpe Ratios of RGOP and GOP VaR: Break-even point 70 years SR: Always 4.24% better on avg. Data Source: Fama French Online Data Library

48 Synthetic Experiment: Ambiguity Effects I Calculate µ and from 0 Industry Portfolios I T = 360 months, = 5% I Compare VaRs of RGOP and GOP under 2 distributions in P outperformance % Lognormal WC-Dist for GOP WC-Dist for RGOP

49 Out-of-Sample Backtests Dataset ˆr p ˆp 0IND 2IND ishares DJIA csr c TR c NR [MDD RGOP /n GOP RGOP /n GOP RGOP /n GOP RGOP /n GOP winners

50 References I Kelly, J. L. A new interpretation of information rate. Bell System Technical Journal 35, 4 (956), I Latanè, H. A. Criteria for choice among risky ventures. Journal of Political Economy 67, 2 (959), I Roy, A. D. Safety first and the holding of assets. Econometrica 20, 3 (952), I Rubinstein., M. Continuously rebalanced investment strategies. Journal of Portfolio Management 8, (99), I Rujeerapaiboon, N., Kuhn, D., and Wiesemann, W. Robust growth-optimal portfolios. Working Paper (204). I Yu, Y., Li, Y., Schuurmans, D., and Szepesvári, C. A general projection property for distribution families. Advances in Neural Information Processing Systems 22 (2009), I Zymler, S., Kuhn, D., and Rustem, B. Distributionally robust joint chance constraints with second-order moment information. Mathematical Programming 37, -2 (203), I Zymler, S., Kuhn, D., and Rustem, B. Worst-case value-at-risk of nonlinear portfolios. Management Science 59, (203),

Distributionally Robust Convex Optimization

Distributionally Robust Convex Optimization Submitted to Operations Research manuscript OPRE-2013-02-060 Authors are encouraged to submit new papers to INFORMS journals by means of a style file template, which includes the journal title. However,

More information

Multi-Period Portfolio Optimization: Translation of Autocorrelation Risk to Excess Variance

Multi-Period Portfolio Optimization: Translation of Autocorrelation Risk to Excess Variance Multi-Period Portfolio Optimization: Translation of Autocorrelation Risk to Excess Variance Byung-Geun Choi a, Napat Rujeerapaiboon b, Ruiwei Jiang a a Department of Industrial & Operations Engineering,

More information

Distributionally Robust Convex Optimization

Distributionally Robust Convex Optimization Distributionally Robust Convex Optimization Wolfram Wiesemann 1, Daniel Kuhn 1, and Melvyn Sim 2 1 Department of Computing, Imperial College London, United Kingdom 2 Department of Decision Sciences, National

More information

Semidefinite and Second Order Cone Programming Seminar Fall 2012 Project: Robust Optimization and its Application of Robust Portfolio Optimization

Semidefinite and Second Order Cone Programming Seminar Fall 2012 Project: Robust Optimization and its Application of Robust Portfolio Optimization Semidefinite and Second Order Cone Programming Seminar Fall 2012 Project: Robust Optimization and its Application of Robust Portfolio Optimization Instructor: Farid Alizadeh Author: Ai Kagawa 12/12/2012

More information

Ambiguous Joint Chance Constraints under Mean and Dispersion Information

Ambiguous Joint Chance Constraints under Mean and Dispersion Information Ambiguous Joint Chance Constraints under Mean and Dispersion Information Grani A. Hanasusanto 1, Vladimir Roitch 2, Daniel Kuhn 3, and Wolfram Wiesemann 4 1 Graduate Program in Operations Research and

More information

Safe Approximations of Chance Constraints Using Historical Data

Safe Approximations of Chance Constraints Using Historical Data Safe Approximations of Chance Constraints Using Historical Data İhsan Yanıkoğlu Department of Econometrics and Operations Research, Tilburg University, 5000 LE, Netherlands, {i.yanikoglu@uvt.nl} Dick den

More information

Data-driven Chance Constrained Stochastic Program

Data-driven Chance Constrained Stochastic Program Data-driven Chance Constrained Stochastic Program Ruiwei Jiang and Yongpei Guan Department of Industrial and Systems Engineering University of Florida, Gainesville, FL 326, USA Email: rwjiang@ufl.edu;

More information

EE 227A: Convex Optimization and Applications April 24, 2008

EE 227A: Convex Optimization and Applications April 24, 2008 EE 227A: Convex Optimization and Applications April 24, 2008 Lecture 24: Robust Optimization: Chance Constraints Lecturer: Laurent El Ghaoui Reading assignment: Chapter 2 of the book on Robust Optimization

More information

Handout 8: Dealing with Data Uncertainty

Handout 8: Dealing with Data Uncertainty MFE 5100: Optimization 2015 16 First Term Handout 8: Dealing with Data Uncertainty Instructor: Anthony Man Cho So December 1, 2015 1 Introduction Conic linear programming CLP, and in particular, semidefinite

More information

Modern Portfolio Theory with Homogeneous Risk Measures

Modern Portfolio Theory with Homogeneous Risk Measures Modern Portfolio Theory with Homogeneous Risk Measures Dirk Tasche Zentrum Mathematik Technische Universität München http://www.ma.tum.de/stat/ Rotterdam February 8, 2001 Abstract The Modern Portfolio

More information

Robust Efficient Frontier Analysis with a Separable Uncertainty Model

Robust Efficient Frontier Analysis with a Separable Uncertainty Model Robust Efficient Frontier Analysis with a Separable Uncertainty Model Seung-Jean Kim Stephen Boyd October 2007 Abstract Mean-variance (MV) analysis is often sensitive to model mis-specification or uncertainty,

More information

Distributionally Robust Discrete Optimization with Entropic Value-at-Risk

Distributionally Robust Discrete Optimization with Entropic Value-at-Risk Distributionally Robust Discrete Optimization with Entropic Value-at-Risk Daniel Zhuoyu Long Department of SEEM, The Chinese University of Hong Kong, zylong@se.cuhk.edu.hk Jin Qi NUS Business School, National

More information

On deterministic reformulations of distributionally robust joint chance constrained optimization problems

On deterministic reformulations of distributionally robust joint chance constrained optimization problems On deterministic reformulations of distributionally robust joint chance constrained optimization problems Weijun Xie and Shabbir Ahmed School of Industrial & Systems Engineering Georgia Institute of Technology,

More information

Birgit Rudloff Operations Research and Financial Engineering, Princeton University

Birgit Rudloff Operations Research and Financial Engineering, Princeton University TIME CONSISTENT RISK AVERSE DYNAMIC DECISION MODELS: AN ECONOMIC INTERPRETATION Birgit Rudloff Operations Research and Financial Engineering, Princeton University brudloff@princeton.edu Alexandre Street

More information

Robust Optimization: Applications in Portfolio Selection Problems

Robust Optimization: Applications in Portfolio Selection Problems Robust Optimization: Applications in Portfolio Selection Problems Vris Cheung and Henry Wolkowicz WatRISQ University of Waterloo Vris Cheung (University of Waterloo) Robust optimization 2009 1 / 19 Outline

More information

Pareto Efficiency in Robust Optimization

Pareto Efficiency in Robust Optimization Pareto Efficiency in Robust Optimization Dan Iancu Graduate School of Business Stanford University joint work with Nikolaos Trichakis (HBS) 1/26 Classical Robust Optimization Typical linear optimization

More information

Robust Optimization for Risk Control in Enterprise-wide Optimization

Robust Optimization for Risk Control in Enterprise-wide Optimization Robust Optimization for Risk Control in Enterprise-wide Optimization Juan Pablo Vielma Department of Industrial Engineering University of Pittsburgh EWO Seminar, 011 Pittsburgh, PA Uncertainty in Optimization

More information

Mathematical Optimization Models and Applications

Mathematical Optimization Models and Applications Mathematical Optimization Models and Applications Yinyu Ye Department of Management Science and Engineering Stanford University Stanford, CA 94305, U.S.A. http://www.stanford.edu/ yyye Chapters 1, 2.1-2,

More information

Distributionally robust simple integer recourse

Distributionally robust simple integer recourse Distributionally robust simple integer recourse Weijun Xie 1 and Shabbir Ahmed 2 1 Department of Industrial and Systems Engineering, Virginia Tech, Blacksburg, VA 24061 2 School of Industrial & Systems

More information

Miloš Kopa. Decision problems with stochastic dominance constraints

Miloš Kopa. Decision problems with stochastic dominance constraints Decision problems with stochastic dominance constraints Motivation Portfolio selection model Mean risk models max λ Λ m(λ r) νr(λ r) or min λ Λ r(λ r) s.t. m(λ r) µ r is a random vector of assets returns

More information

Robust Markowitz portfolio selection. ambiguous covariance matrix

Robust Markowitz portfolio selection. ambiguous covariance matrix under ambiguous covariance matrix University Paris Diderot, LPMA Sorbonne Paris Cité Based on joint work with A. Ismail, Natixis MFO March 2, 2017 Outline Introduction 1 Introduction 2 3 and Sharpe ratio

More information

Distributionally Robust Stochastic Optimization with Wasserstein Distance

Distributionally Robust Stochastic Optimization with Wasserstein Distance Distributionally Robust Stochastic Optimization with Wasserstein Distance Rui Gao DOS Seminar, Oct 2016 Joint work with Anton Kleywegt School of Industrial and Systems Engineering Georgia Tech What is

More information

Optimized Bonferroni Approximations of Distributionally Robust Joint Chance Constraints

Optimized Bonferroni Approximations of Distributionally Robust Joint Chance Constraints Optimized Bonferroni Approximations of Distributionally Robust Joint Chance Constraints Weijun Xie Shabbir Ahmed Ruiwei Jiang February 13, 2017 Abstract A distributionally robust joint chance constraint

More information

c 2014 Society for Industrial and Applied Mathematics

c 2014 Society for Industrial and Applied Mathematics SIAM J. OPTIM. Vol. 4, No. 3, pp. 1485 1506 c 014 Society for Industrial and Applied Mathematics DISTRIBUTIONALLY ROBUST STOCHASTIC KNAPSACK PROBLEM JIANQIANG CHENG, ERICK DELAGE, AND ABDEL LISSER Abstract.

More information

On distributional robust probability functions and their computations

On distributional robust probability functions and their computations On distributional robust probability functions and their computations Man Hong WONG a,, Shuzhong ZHANG b a Department of Systems Engineering and Engineering Management, The Chinese University of Hong Kong,

More information

Portfolio Selection under Model Uncertainty:

Portfolio Selection under Model Uncertainty: manuscript No. (will be inserted by the editor) Portfolio Selection under Model Uncertainty: A Penalized Moment-Based Optimization Approach Jonathan Y. Li Roy H. Kwon Received: date / Accepted: date Abstract

More information

Data-Driven Optimization under Distributional Uncertainty

Data-Driven Optimization under Distributional Uncertainty Data-Driven Optimization under Distributional Uncertainty Postdoctoral Researcher Electrical and Systems Engineering University of Pennsylvania A network of physical objects - Devices - Vehicles - Buildings

More information

Robust portfolio selection under norm uncertainty

Robust portfolio selection under norm uncertainty Wang and Cheng Journal of Inequalities and Applications (2016) 2016:164 DOI 10.1186/s13660-016-1102-4 R E S E A R C H Open Access Robust portfolio selection under norm uncertainty Lei Wang 1 and Xi Cheng

More information

Convex optimization problems. Optimization problem in standard form

Convex optimization problems. Optimization problem in standard form Convex optimization problems optimization problem in standard form convex optimization problems linear optimization quadratic optimization geometric programming quasiconvex optimization generalized inequality

More information

Distributionally Robust Optimization under Moment Uncertainty with Application to Data-Driven Problems

Distributionally Robust Optimization under Moment Uncertainty with Application to Data-Driven Problems OPERATIONS RESEARCH Vol. 00, No. 0, Xxxxx 0000, pp. 000 000 ISSN 0030-364X EISSN 526-5463 00 0000 000 INFORMS DOI 0.287/xxxx.0000.0000 c 0000 INFORMS Distributionally Robust Optimization under Moment Uncertainty

More information

Random Convex Approximations of Ambiguous Chance Constrained Programs

Random Convex Approximations of Ambiguous Chance Constrained Programs Random Convex Approximations of Ambiguous Chance Constrained Programs Shih-Hao Tseng Eilyan Bitar Ao Tang Abstract We investigate an approach to the approximation of ambiguous chance constrained programs

More information

Motivation General concept of CVaR Optimization Comparison. VaR and CVaR. Přemysl Bejda.

Motivation General concept of CVaR Optimization Comparison. VaR and CVaR. Přemysl Bejda. VaR and CVaR Přemysl Bejda premyslbejda@gmail.com 2014 Contents 1 Motivation 2 General concept of CVaR 3 Optimization 4 Comparison Contents 1 Motivation 2 General concept of CVaR 3 Optimization 4 Comparison

More information

Data-Driven Distributionally Robust Chance-Constrained Optimization with Wasserstein Metric

Data-Driven Distributionally Robust Chance-Constrained Optimization with Wasserstein Metric Data-Driven Distributionally Robust Chance-Constrained Optimization with asserstein Metric Ran Ji Department of System Engineering and Operations Research, George Mason University, rji2@gmu.edu; Miguel

More information

Central-limit approach to risk-aware Markov decision processes

Central-limit approach to risk-aware Markov decision processes Central-limit approach to risk-aware Markov decision processes Jia Yuan Yu Concordia University November 27, 2015 Joint work with Pengqian Yu and Huan Xu. Inventory Management 1 1 Look at current inventory

More information

Handout 6: Some Applications of Conic Linear Programming

Handout 6: Some Applications of Conic Linear Programming ENGG 550: Foundations of Optimization 08 9 First Term Handout 6: Some Applications of Conic Linear Programming Instructor: Anthony Man Cho So November, 08 Introduction Conic linear programming CLP, and

More information

Estimation and Optimization: Gaps and Bridges. MURI Meeting June 20, Laurent El Ghaoui. UC Berkeley EECS

Estimation and Optimization: Gaps and Bridges. MURI Meeting June 20, Laurent El Ghaoui. UC Berkeley EECS MURI Meeting June 20, 2001 Estimation and Optimization: Gaps and Bridges Laurent El Ghaoui EECS UC Berkeley 1 goals currently, estimation (of model parameters) and optimization (of decision variables)

More information

Distributionally robust optimization techniques in batch bayesian optimisation

Distributionally robust optimization techniques in batch bayesian optimisation Distributionally robust optimization techniques in batch bayesian optimisation Nikitas Rontsis June 13, 2016 1 Introduction This report is concerned with performing batch bayesian optimization of an unknown

More information

Distributionally robust optimization with polynomial densities: theory, models and algorithms

Distributionally robust optimization with polynomial densities: theory, models and algorithms Distributionally robust optimization with polynomial densities: theory, models and algorithms Etienne de lerk Daniel uhn rzysztof Postek Abstract In distributionally robust optimization the probability

More information

Tractable Robust Expected Utility and Risk Models for Portfolio Optimization

Tractable Robust Expected Utility and Risk Models for Portfolio Optimization Tractable Robust Expected Utility and Risk Models for Portfolio Optimization Karthik Natarajan Melvyn Sim Joline Uichanco Submitted: March 13, 2008. Revised: September 25, 2008, December 5, 2008 Abstract

More information

Distributionally robust control of constrained. stochastic systems

Distributionally robust control of constrained. stochastic systems Distributionally robust control of constrained 1 stochastic systems Bart P.G. Van Parys, Daniel Kuhn, Paul J. Goulart and Manfred Morari Abstract We investigate the control of constrained stochastic linear

More information

MFM Practitioner Module: Risk & Asset Allocation. John Dodson. January 25, 2012

MFM Practitioner Module: Risk & Asset Allocation. John Dodson. January 25, 2012 MFM Practitioner Module: Risk & Asset Allocation January 25, 2012 Optimizing Allocations Once we have 1. chosen the markets and an investment horizon 2. modeled the markets 3. agreed on an objective with

More information

Thomas Knispel Leibniz Universität Hannover

Thomas Knispel Leibniz Universität Hannover Optimal long term investment under model ambiguity Optimal long term investment under model ambiguity homas Knispel Leibniz Universität Hannover knispel@stochastik.uni-hannover.de AnStAp0 Vienna, July

More information

Stability Analysis for Mathematical Programs with Distributionally Robust Chance Constraint

Stability Analysis for Mathematical Programs with Distributionally Robust Chance Constraint Stability Analysis for Mathematical Programs with Distributionally Robust Chance Constraint Shaoyan Guo, Huifu Xu and Liwei Zhang September 24, 2015 Abstract. Stability analysis for optimization problems

More information

Convex Optimization in Classification Problems

Convex Optimization in Classification Problems New Trends in Optimization and Computational Algorithms December 9 13, 2001 Convex Optimization in Classification Problems Laurent El Ghaoui Department of EECS, UC Berkeley elghaoui@eecs.berkeley.edu 1

More information

arxiv: v3 [math.oc] 25 Apr 2018

arxiv: v3 [math.oc] 25 Apr 2018 Problem-driven scenario generation: an analytical approach for stochastic programs with tail risk measure Jamie Fairbrother *, Amanda Turner *, and Stein W. Wallace ** * STOR-i Centre for Doctoral Training,

More information

4. Convex optimization problems

4. Convex optimization problems Convex Optimization Boyd & Vandenberghe 4. Convex optimization problems optimization problem in standard form convex optimization problems quasiconvex optimization linear optimization quadratic optimization

More information

Chance constrained optimization - applications, properties and numerical issues

Chance constrained optimization - applications, properties and numerical issues Chance constrained optimization - applications, properties and numerical issues Dr. Abebe Geletu Ilmenau University of Technology Department of Simulation and Optimal Processes (SOP) May 31, 2012 This

More information

Reformulation of chance constrained problems using penalty functions

Reformulation of chance constrained problems using penalty functions Reformulation of chance constrained problems using penalty functions Martin Branda Charles University in Prague Faculty of Mathematics and Physics EURO XXIV July 11-14, 2010, Lisbon Martin Branda (MFF

More information

Optimization Problems with Probabilistic Constraints

Optimization Problems with Probabilistic Constraints Optimization Problems with Probabilistic Constraints R. Henrion Weierstrass Institute Berlin 10 th International Conference on Stochastic Programming University of Arizona, Tucson Recommended Reading A.

More information

Quantifying Stochastic Model Errors via Robust Optimization

Quantifying Stochastic Model Errors via Robust Optimization Quantifying Stochastic Model Errors via Robust Optimization IPAM Workshop on Uncertainty Quantification for Multiscale Stochastic Systems and Applications Jan 19, 2016 Henry Lam Industrial & Operations

More information

Sequential Convex Approximations to Joint Chance Constrained Programs: A Monte Carlo Approach

Sequential Convex Approximations to Joint Chance Constrained Programs: A Monte Carlo Approach Sequential Convex Approximations to Joint Chance Constrained Programs: A Monte Carlo Approach L. Jeff Hong Department of Industrial Engineering and Logistics Management The Hong Kong University of Science

More information

Lecture 1: Introduction

Lecture 1: Introduction EE 227A: Convex Optimization and Applications January 17 Lecture 1: Introduction Lecturer: Anh Pham Reading assignment: Chapter 1 of BV 1. Course outline and organization Course web page: http://www.eecs.berkeley.edu/~elghaoui/teaching/ee227a/

More information

WORST-CASE VALUE-AT-RISK AND ROBUST PORTFOLIO OPTIMIZATION: A CONIC PROGRAMMING APPROACH

WORST-CASE VALUE-AT-RISK AND ROBUST PORTFOLIO OPTIMIZATION: A CONIC PROGRAMMING APPROACH WORST-CASE VALUE-AT-RISK AND ROBUST PORTFOLIO OPTIMIZATION: A CONIC PROGRAMMING APPROACH LAURENT EL GHAOUI Department of Electrical Engineering and Computer Sciences, University of California, Berkeley,

More information

Robust Markov Decision Processes

Robust Markov Decision Processes Robust Markov Decision Processes Wolfram Wiesemann, Daniel Kuhn and Berç Rustem February 9, 2012 Abstract Markov decision processes (MDPs) are powerful tools for decision making in uncertain dynamic environments.

More information

On Distributionally Robust Chance Constrained Program with Wasserstein Distance

On Distributionally Robust Chance Constrained Program with Wasserstein Distance On Distributionally Robust Chance Constrained Program with Wasserstein Distance Weijun Xie 1 1 Department of Industrial and Systems Engineering, Virginia Tech, Blacksburg, VA 4061 June 15, 018 Abstract

More information

Stochastic Optimization with Risk Measures

Stochastic Optimization with Risk Measures Stochastic Optimization with Risk Measures IMA New Directions Short Course on Mathematical Optimization Jim Luedtke Department of Industrial and Systems Engineering University of Wisconsin-Madison August

More information

A SECOND ORDER STOCHASTIC DOMINANCE PORTFOLIO EFFICIENCY MEASURE

A SECOND ORDER STOCHASTIC DOMINANCE PORTFOLIO EFFICIENCY MEASURE K Y B E R N E I K A V O L U M E 4 4 ( 2 0 0 8 ), N U M B E R 2, P A G E S 2 4 3 2 5 8 A SECOND ORDER SOCHASIC DOMINANCE PORFOLIO EFFICIENCY MEASURE Miloš Kopa and Petr Chovanec In this paper, we introduce

More information

Robustness and bootstrap techniques in portfolio efficiency tests

Robustness and bootstrap techniques in portfolio efficiency tests Robustness and bootstrap techniques in portfolio efficiency tests Dept. of Probability and Mathematical Statistics, Charles University, Prague, Czech Republic July 8, 2013 Motivation Portfolio selection

More information

I.3. LMI DUALITY. Didier HENRION EECI Graduate School on Control Supélec - Spring 2010

I.3. LMI DUALITY. Didier HENRION EECI Graduate School on Control Supélec - Spring 2010 I.3. LMI DUALITY Didier HENRION henrion@laas.fr EECI Graduate School on Control Supélec - Spring 2010 Primal and dual For primal problem p = inf x g 0 (x) s.t. g i (x) 0 define Lagrangian L(x, z) = g 0

More information

Distributionally Robust Optimization with ROME (part 1)

Distributionally Robust Optimization with ROME (part 1) Distributionally Robust Optimization with ROME (part 1) Joel Goh Melvyn Sim Department of Decision Sciences NUS Business School, Singapore 18 Jun 2009 NUS Business School Guest Lecture J. Goh, M. Sim (NUS)

More information

Optimization Tools in an Uncertain Environment

Optimization Tools in an Uncertain Environment Optimization Tools in an Uncertain Environment Michael C. Ferris University of Wisconsin, Madison Uncertainty Workshop, Chicago: July 21, 2008 Michael Ferris (University of Wisconsin) Stochastic optimization

More information

Distributionally Robust Multi-Item Newsvendor Problems with Multimodal Demand Distributions

Distributionally Robust Multi-Item Newsvendor Problems with Multimodal Demand Distributions Distributionally Robust Multi-Item Newsvendor Problems with Multimodal Demand Distributions Grani A. Hanasusanto, Daniel Kuhn, Stein W. Wallace 3, and Steve Zymler Department of Computing, Imperial College

More information

THE stochastic and dynamic environments of many practical

THE stochastic and dynamic environments of many practical A Convex Optimization Approach to Distributionally Robust Markov Decision Processes with Wasserstein Distance Insoon Yang, Member, IEEE Abstract In this paper, we consider the problem of constructing control

More information

ORIGINS OF STOCHASTIC PROGRAMMING

ORIGINS OF STOCHASTIC PROGRAMMING ORIGINS OF STOCHASTIC PROGRAMMING Early 1950 s: in applications of Linear Programming unknown values of coefficients: demands, technological coefficients, yields, etc. QUOTATION Dantzig, Interfaces 20,1990

More information

Models and Algorithms for Distributionally Robust Least Squares Problems

Models and Algorithms for Distributionally Robust Least Squares Problems Models and Algorithms for Distributionally Robust Least Squares Problems Sanjay Mehrotra and He Zhang February 12, 2011 Abstract We present different robust frameworks using probabilistic ambiguity descriptions

More information

Convex Optimization and l 1 -minimization

Convex Optimization and l 1 -minimization Convex Optimization and l 1 -minimization Sangwoon Yun Computational Sciences Korea Institute for Advanced Study December 11, 2009 2009 NIMS Thematic Winter School Outline I. Convex Optimization II. l

More information

Optimized Bonferroni Approximations of Distributionally Robust Joint Chance Constraints

Optimized Bonferroni Approximations of Distributionally Robust Joint Chance Constraints Optimized Bonferroni Approximations of Distributionally Robust Joint Chance Constraints Weijun Xie 1, Shabbir Ahmed 2, Ruiwei Jiang 3 1 Department of Industrial and Systems Engineering, Virginia Tech,

More information

Generalized Hypothesis Testing and Maximizing the Success Probability in Financial Markets

Generalized Hypothesis Testing and Maximizing the Success Probability in Financial Markets Generalized Hypothesis Testing and Maximizing the Success Probability in Financial Markets Tim Leung 1, Qingshuo Song 2, and Jie Yang 3 1 Columbia University, New York, USA; leung@ieor.columbia.edu 2 City

More information

Tractable Robust Expected Utility and Risk Models for Portfolio Optimization

Tractable Robust Expected Utility and Risk Models for Portfolio Optimization Tractable Robust Expected Utility and Risk Models for Portfolio Optimization Karthik Natarajan Melvyn Sim Joline Uichanco Submitted: March 13, 2008 Abstract Expected utility models in portfolio optimization

More information

Scenario estimation and generation

Scenario estimation and generation October 10, 2004 The one-period case Distances of Probability Measures Tensor products of trees Tree reduction A decision problem is subject to uncertainty Uncertainty is represented by probability To

More information

Selected Examples of CONIC DUALITY AT WORK Robust Linear Optimization Synthesis of Linear Controllers Matrix Cube Theorem A.

Selected Examples of CONIC DUALITY AT WORK Robust Linear Optimization Synthesis of Linear Controllers Matrix Cube Theorem A. . Selected Examples of CONIC DUALITY AT WORK Robust Linear Optimization Synthesis of Linear Controllers Matrix Cube Theorem A. Nemirovski Arkadi.Nemirovski@isye.gatech.edu Linear Optimization Problem,

More information

January 29, Introduction to optimization and complexity. Outline. Introduction. Problem formulation. Convexity reminder. Optimality Conditions

January 29, Introduction to optimization and complexity. Outline. Introduction. Problem formulation. Convexity reminder. Optimality Conditions Olga Galinina olga.galinina@tut.fi ELT-53656 Network Analysis Dimensioning II Department of Electronics Communications Engineering Tampere University of Technology, Tampere, Finl January 29, 2014 1 2 3

More information

approximation algorithms I

approximation algorithms I SUM-OF-SQUARES method and approximation algorithms I David Steurer Cornell Cargese Workshop, 201 meta-task encoded as low-degree polynomial in R x example: f(x) = i,j n w ij x i x j 2 given: functions

More information

Quadratic Two-Stage Stochastic Optimization with Coherent Measures of Risk

Quadratic Two-Stage Stochastic Optimization with Coherent Measures of Risk Noname manuscript No. (will be inserted by the editor) Quadratic Two-Stage Stochastic Optimization with Coherent Measures of Risk Jie Sun Li-Zhi Liao Brian Rodrigues Received: date / Accepted: date Abstract

More information

Chance constrained optimization

Chance constrained optimization Chance constrained optimization chance constraints and percentile optimization chance constraints for log-concave distributions convex approximation of chance constraints sources: Rockafellar & Uryasev,

More information

On optimal quadratic Lyapunov functions for polynomial systems

On optimal quadratic Lyapunov functions for polynomial systems On optimal quadratic Lyapunov functions for polynomial systems G. Chesi 1,A.Tesi 2, A. Vicino 1 1 Dipartimento di Ingegneria dell Informazione, Università disiena Via Roma 56, 53100 Siena, Italy 2 Dipartimento

More information

Scenario Generation and Sampling Methods

Scenario Generation and Sampling Methods Scenario Generation and Sampling Methods Güzin Bayraksan Tito Homem-de-Mello SVAN 2016 IMPA May 11th, 2016 Bayraksan (OSU) & Homem-de-Mello (UAI) Scenario Generation and Sampling SVAN IMPA May 11 1 / 33

More information

CVaR and Examples of Deviation Risk Measures

CVaR and Examples of Deviation Risk Measures CVaR and Examples of Deviation Risk Measures Jakub Černý Department of Probability and Mathematical Statistics Stochastic Modelling in Economics and Finance November 10, 2014 1 / 25 Contents CVaR - Dual

More information

Distributionally Robust Reward-risk Ratio Programming with Wasserstein Metric

Distributionally Robust Reward-risk Ratio Programming with Wasserstein Metric oname manuscript o. will be inserted by the editor) Distributionally Robust Reward-risk Ratio Programming with Wasserstein Metric Yong Zhao Yongchao Liu Jin Zhang Xinmin Yang Received: date / Accepted:

More information

Research Article Optimal Portfolio Estimation for Dependent Financial Returns with Generalized Empirical Likelihood

Research Article Optimal Portfolio Estimation for Dependent Financial Returns with Generalized Empirical Likelihood Advances in Decision Sciences Volume 2012, Article ID 973173, 8 pages doi:10.1155/2012/973173 Research Article Optimal Portfolio Estimation for Dependent Financial Returns with Generalized Empirical Likelihood

More information

Optimizing allocations

Optimizing allocations 6 Optimizing allocations In this chapter we determine the optimal allocation for a generic investor in a generic market of securities. In Section 6.1 we introduce allocation optimization by means of a

More information

Robust Fragmentation: A Data-Driven Approach to Decision-Making under Distributional Ambiguity

Robust Fragmentation: A Data-Driven Approach to Decision-Making under Distributional Ambiguity Robust Fragmentation: A Data-Driven Approach to Decision-Making under Distributional Ambiguity A THESIS SUBMITTED TO THE FACULTY OF THE GRADUATE SCHOOL OF THE UNIVERSITY OF MINNESOTA BY Jeffrey Moulton

More information

IE598 Big Data Optimization Introduction

IE598 Big Data Optimization Introduction IE598 Big Data Optimization Introduction Instructor: Niao He Jan 17, 2018 1 A little about me Assistant Professor, ISE & CSL UIUC, 2016 Ph.D. in Operations Research, M.S. in Computational Sci. & Eng. Georgia

More information

Histogram Models for Robust Portfolio Optimization. Daniel Bienstock January 2007, revised July 2007

Histogram Models for Robust Portfolio Optimization. Daniel Bienstock January 2007, revised July 2007 Center for Financial Engineering, Columbia University Histogram Models for Robust Portfolio Optimization Daniel Bienstock January 2007, revised July 2007 Abstract We present experimental results on portfolio

More information

Sparse and Robust Optimization and Applications

Sparse and Robust Optimization and Applications Sparse and and Statistical Learning Workshop Les Houches, 2013 Robust Laurent El Ghaoui with Mert Pilanci, Anh Pham EECS Dept., UC Berkeley January 7, 2013 1 / 36 Outline Sparse Sparse Sparse Probability

More information

Worst-Case Conditional Value-at-Risk with Application to Robust Portfolio Management

Worst-Case Conditional Value-at-Risk with Application to Robust Portfolio Management Worst-Case Conditional Value-at-Risk with Application to Robust Portfolio Management Shu-Shang Zhu Department of Management Science, School of Management, Fudan University, Shanghai 200433, China, sszhu@fudan.edu.cn

More information

CVAR REDUCED FUZZY VARIABLES AND THEIR SECOND ORDER MOMENTS

CVAR REDUCED FUZZY VARIABLES AND THEIR SECOND ORDER MOMENTS Iranian Journal of Fuzzy Systems Vol., No. 5, (05 pp. 45-75 45 CVAR REDUCED FUZZY VARIABLES AND THEIR SECOND ORDER MOMENTS X. J. BAI AND Y. K. LIU Abstract. Based on credibilistic value-at-risk (CVaR of

More information

Portfolio optimization with stochastic dominance constraints

Portfolio optimization with stochastic dominance constraints Charles University in Prague Faculty of Mathematics and Physics Portfolio optimization with stochastic dominance constraints December 16, 2014 Contents Motivation 1 Motivation 2 3 4 5 Contents Motivation

More information

DISTRIBUTIONALLY ROBUST STOCHASTIC KNAPSACK PROBLEM

DISTRIBUTIONALLY ROBUST STOCHASTIC KNAPSACK PROBLEM DISTRIBUTIONALLY ROBUST STOCHASTIC KNAPSACK PROBLEM JIANQIANG CHENG, ERICK DELAGE, AND ABDEL LISSER Abstract. This paper considers a distributionally robust version of a quadratic knapsack problem. In

More information

University Of Calgary Department of Mathematics and Statistics

University Of Calgary Department of Mathematics and Statistics University Of Calgary Department of Mathematics and Statistics Hawkes Seminar May 23, 2018 1 / 46 Some Problems in Insurance and Reinsurance Mohamed Badaoui Department of Electrical Engineering National

More information

Near-Optimal Linear Recovery from Indirect Observations

Near-Optimal Linear Recovery from Indirect Observations Near-Optimal Linear Recovery from Indirect Observations joint work with A. Nemirovski, Georgia Tech http://www2.isye.gatech.edu/ nemirovs/statopt LN.pdf Les Houches, April, 2017 Situation: In the nature

More information

Robust Quadratic Programming with Mixed-Integer Uncertainty

Robust Quadratic Programming with Mixed-Integer Uncertainty Robust Quadratic Programming with Mixed-Integer Uncertainty Areesh Mittal, Can Gokalp, and Grani A. Hanasusanto arxiv:706.0949v7 [math.oc] 8 Dec 08 Graduate Program in Operations Research and Industrial

More information

Scenario-Free Stochastic Programming

Scenario-Free Stochastic Programming Scenario-Free Stochastic Programming Wolfram Wiesemann, Angelos Georghiou, and Daniel Kuhn Department of Computing Imperial College London London SW7 2AZ, United Kingdom December 17, 2010 Outline 1 Deterministic

More information

Robust portfolio selection based on a joint ellipsoidal uncertainty set

Robust portfolio selection based on a joint ellipsoidal uncertainty set Optimization Methods & Software Vol. 00, No. 0, Month 2009, 1 16 Robust portfolio selection based on a joint ellipsoidal uncertainty set Zhaosong Lu* Department of Mathematics, Simon Fraser University,

More information

Multi-Range Robust Optimization vs Stochastic Programming in Prioritizing Project Selection

Multi-Range Robust Optimization vs Stochastic Programming in Prioritizing Project Selection Multi-Range Robust Optimization vs Stochastic Programming in Prioritizing Project Selection Ruken Düzgün Aurélie Thiele July 2012 Abstract This paper describes a multi-range robust optimization approach

More information

Convex Optimization M2

Convex Optimization M2 Convex Optimization M2 Lecture 3 A. d Aspremont. Convex Optimization M2. 1/49 Duality A. d Aspremont. Convex Optimization M2. 2/49 DMs DM par email: dm.daspremont@gmail.com A. d Aspremont. Convex Optimization

More information

CORC Technical Report TR Robust portfolio selection problems

CORC Technical Report TR Robust portfolio selection problems CORC Technical Report TR-2002-03 Robust portfolio selection problems D. Goldfarb G. Iyengar Submitted: Dec. 26, 200. Revised: May 25th, 2002, July 24th, 2002 Abstract In this paper we show how to formulate

More information

A Robust Portfolio Technique for Mitigating the Fragility of CVaR Minimization

A Robust Portfolio Technique for Mitigating the Fragility of CVaR Minimization A Robust Portfolio Technique for Mitigating the Fragility of CVaR Minimization Jun-ya Gotoh Department of Industrial and Systems Engineering Chuo University 2-3-27 Kasuga, Bunkyo-ku, Tokyo 2-855, Japan

More information

Likelihood robust optimization for data-driven problems

Likelihood robust optimization for data-driven problems Comput Manag Sci (2016) 13:241 261 DOI 101007/s10287-015-0240-3 ORIGINAL PAPER Likelihood robust optimization for data-driven problems Zizhuo Wang 1 Peter W Glynn 2 Yinyu Ye 2 Received: 9 August 2013 /

More information

4. Convex optimization problems

4. Convex optimization problems Convex Optimization Boyd & Vandenberghe 4. Convex optimization problems optimization problem in standard form convex optimization problems quasiconvex optimization linear optimization quadratic optimization

More information