Regularizing portfolio optimization

The optimization of large portfolios displays an inherent instability due to estimation error. This poses a fundamental problem, because solutions that are not stable under sample fluctuations may look optimal for a given sample, but are, in effect, very far from optimal with respect to the average risk. In this paper, we approach the problem from the point of view of statistical learning theory. The occurrence of the instability is intimately related to over-fitting, which can be avoided using known regularization methods. We show how regularized portfolio optimization with the expected shortfall as a risk measure is related to support vector regression. The budget constraint dictates a modification. We present the resulting optimization problem and discuss the solution. The L2 norm of the weight vector is used as a regularizer, which corresponds to a diversification ‘pressure’. This means that diversification, besides counteracting downward fluctuations in some assets by upward fluctuations in others, is also crucial because it improves the stability of the solution. The approach we provide here allows for the simultaneous treatment of optimization and diversification in one framework that enables the investor to trade off between the two, depending on the size of the available dataset.

[1]  Imre Kondor,et al.  Instability of Portfolio Optimization under Coherent Risk Measures , 2010, Adv. Complex Syst..

[2]  Raman Uppal,et al.  A Generalized Approach to Portfolio Optimization: Improving Performance by Constraining Portfolio Norms , 2009, Manag. Sci..

[3]  Victor DeMiguel,et al.  Optimal Versus Naive Diversification: How Inefficient is the 1/N Portfolio Strategy? , 2009 .

[4]  Gabriel Frahm,et al.  Dominating Estimators for the Global Minimum Variance Portfolio , 2009, SSRN Electronic Journal.

[5]  I. Kondor,et al.  The instability of downside risk measures , 2008, 0811.0800.

[6]  Gabriel Frahm.  Linear statistical inference for global and local minimum variance portfolios , 2008 .

[7]  Akiko Takeda,et al.  ν-support vector machine as conditional value-at-risk minimization , 2008, ICML '08.

[8]  I. Kondor,et al.  Feasibility of Portfolio Optimization under Coherent Risk Measures , 2008, 0803.2283.

[9]  Imre Kondor,et al.  Noise sensitivity of portfolio selection in constant conditional correlation GARCH models , 2007 .

[10]  Raymond Kan,et al.  Optimal Portfolio Choice with Parameter Uncertainty , 2007, Journal of Financial and Quantitative Analysis.

[11]  Yarema Okhrin,et al.  Multivariate Shrinkage for Optimal Portfolio Weights , 2007 .

[12]  Marc Mézard,et al.  Risk minimization through portfolio replication , 2006, physics/0608035.

[13]  I. Kondor,et al.  Portfolio instability and linear constraints , 2007 .

[14]  J. Weston,et al.  Support Vector Machine Solvers , 2007 .

[15]  I. Kondor,et al.  Noise sensitivity of portfolio selection under various risk measures , 2006, physics/0611027.

[16]  Christoph Memmel,et al.  Estimating the Global Minimum Variance Portfolio , 2006 .

[17]  Yarema Okhrin,et al.  Distributional properties of portfolio weights , 2006 .

[18]  Marc Mézard,et al.  On the feasibility of portfolio optimization under expected shortfall , 2006, physics/0606015.

[19]  Jason Weston,et al.  Fast Kernel Classifiers with Online and Active Learning , 2005, J. Mach. Learn. Res..

[20]  J. Bouchaud,et al.  Financial Applications of Random Matrix Theory: Old Laces and New Pieces , 2005, physics/0507111.

[21]  Bernd Scherer,et al.  Introduction to modern portfolio optimization with NUOPT and S-PLUS , 2005 .

[22]  Lorenzo Garlappi,et al.  Portfolio Selection with Parameter and Model Uncertainty: A Multi-Prior Approach , 2004 .

[23]  Olivier Ledoit,et al.  A well-conditioned estimator for large-dimensional covariance matrices , 2004 .

[24]  Z. Burda,et al.  Signal and Noise in Correlation Matrix , 2003, cond-mat/0305627.

[25]  Imre Kondor,et al.  Estimated correlation matrices and portfolio optimization , 2003, cond-mat/0305475.

[26]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[27]  Yoshua Bengio,et al.  No Unbiased Estimator of the Variance of K-Fold Cross-Validation , 2003, J. Mach. Learn. Res..

[28]  Olivier Ledoit,et al.  Improved estimation of the covariance matrix of stock returns with an application to portfolio selection , 2003 .

[29]  Olivier Ledoit,et al.  Honey, I Shrunk the Sample Covariance Matrix , 2003 .

[30]  Z. Burda,et al.  Is Econophysics a Solid Science , 2003, cond-mat/0301096.

[31]  I. Kondor,et al.  Noisy Covariance Matrices and Portfolio Optimization II , 2002, cond-mat/0205119.

[32]  Bernhard Schölkopf,et al.  Extension of the nu-SVM range for classification , 2003 .

[33]  R. Jagannathan,et al.  Risk Reduction in Large Portfolios: Why Imposing the Wrong Constraints Helps , 2002 .

[34]  I. Kondor,et al.  Noisy covariance matrices and portfolio optimization , 2001, cond-mat/0111503.

[35]  V. Plerou,et al.  Random matrix approach to cross correlations in financial data. , 2001, Physical review. E, Statistical, nonlinear, and soft matter physics.

[36]  Abaxbank,et al.  Spectral Measures of Risk : a Coherent Representation of Subjective Risk Aversion , 2002 .

[37]  C. Acerbi,et al.  On the coherence of expected shortfall , 2001, cond-mat/0104295.

[38]  C. Acerbi,et al.  Expected Shortfall as a Tool for Financial Risk Management , 2001, cond-mat/0102304.

[39]  J. Bouchaud,et al.  RANDOM MATRIX THEORY AND FINANCIAL CORRELATIONS , 2000 .

[40]  Bernhard Schölkopf,et al.  New Support Vector Algorithms , 2000, Neural Computation.

[41]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[42]  J. Bouchaud,et al.  Theory of financial risks : from statistical physics to risk management , 2000 .

[43]  Phhilippe Jorion.  Value at Risk: The New Benchmark for Managing Financial Risk , 2000 .

[44]  R. Rockafellar,et al.  Optimization of conditional value-at risk , 2000 .

[45]  Philippe Artzner,et al.  Coherent Measures of Risk , 1999 .

[46]  V. Plerou,et al.  Universal and Nonuniversal Properties of Cross Correlations in Financial Time Series , 1999, cond-mat/9902283.

[47]  B. Schölkopf,et al.  Advances in kernel methods: support vector learning , 1999 .

[48]  J. Bouchaud,et al.  Noise Dressing of Financial Correlation Matrices , 1998, cond-mat/9810255.

[49]  Nello Cristianini,et al.  Advances in Kernel Methods - Support Vector Learning , 1999 .

[50]  Vladimir Vapnik,et al.  Statistical learning theory , 1998 .

[51]  Bernhard Schölkopf,et al.  Support vector learning , 1997 .

[52]  R. Tibshirani.  Regression Shrinkage and Selection via the Lasso , 1996 .

[53]  Vladimir Naumovich Vapni.  The Nature of Statistical Learning Theory , 1995 .

[54]  J. Friedman,et al.  A Statistical View of Some Chemometrics Regression Tools , 1993 .

[55]  W. Ziemba,et al.  The Effect of Errors in Means, Variances, and Covariances on Optimal Portfolio Choice , 1993 .

[56]  Bernhard E. Boser,et al.  A training algorithm for optimal margin classifiers , 1992, COLT '92.

[57]  Peter A. Frost,et al.  An Empirical Bayes Approach to Efficient Portfolio Selection , 1986, Journal of Financial and Quantitative Analysis.

[58]  Philippe Jorion.  Bayes-Stein Estimation for Portfolio Analysis , 1986, Journal of Financial and Quantitative Analysis.

[59]  E. Elton.  Modern portfolio theory and investment analysis , 1981 .

[60]  R. C. Merton,et al.  On Estimating the Expected Return on the Market: An Exploratory Investigation , 1980 .

[61]  Vladimir Vapnik,et al.  Chervonenkis: On the uniform convergence of relative frequencies of events to their probabilities , 1971 .

[62]  Shun-ichi Amari,et al.  A Theory of Pattern Recognition , 1968 .

[63]  A. Stuart,et al.  Portfolio Selection: Efficient Diversification of Investments , 1959 .

[64]  K. Schittkowski,et al.  NONLINEAR PROGRAMMING , 2022 .