发帖
雷达卡
by Richard O. Michaud
| |
[size=120%]Efficient Asset Management: A Practical Guide to Stock Portfolio Optimization and Asset Allocation Second Edition By Richard O. Michaud and Robert O. Michaud 1 Oxford University Press, Inc., publishes works that further Oxford University’s objective of excellence in research, scholarship, and education. Oxford New York Auckland Cape Town Dar es Salaam Hong Kong Karachi Kuala Lumpur Madrid Melbourne Mexico City Nairobi New Delhi Shanghai Taipei Toronto With offi ces in Argentina Austria Brazil Chile Czech Republic France Greece Guatemala Hungary Italy Japan Poland Portugal Singapore South Korea Switzerland Thailand Turkey Ukraine Vietnam Copyright © 2008 by Oxford University Press, Inc. Published by Oxford University Press, Inc. 198 Madison Avenue, New York, New York 10016 Amazon.com Review: In spite of theoretical benefits, Markowitz mean-variance (MV)optimized portfolios often fail to meet practical investment goals ofmarketability, usability, and performance, prompting many investors toseek simpler alternatives. Financial experts Richard and Robert Michaud demonstrate that the limitations of MV optimization are not the result of conceptualflaws in Markowitz theory but unrealistic representation of investmentinformation. What is missing is a realistic treatment of estimationerror in the optimization and rebalancing process. The text provides a non-technical review of classical Markowitzoptimization and traditional objections. The authors demonstrate thatin practice the single most important limitation of MV optimization isoversensitivity to estimation error. Portfolio optimization requires amodern statistical perspective. Efficient Asset Management, Second Editionuses Monte Carlo resampling to address information uncertainty anddefine Resampled Efficiency(TM) (RE) technology. RE optimizedportfolios represent a new definition of portfolio optimality that ismore investment intuitive, robust, and provably investment effective. RE rebalancing provides the firstrigorous portfolio trading, monitoring, and asset importance rules,avoiding widespread ad hoc methods in current practice. The Second Edition resolves several open issues and misunderstandingsthat have emerged since the original edition. The new edition includesnew proofs of effectiveness, substantial revisions of statisticalestimation, extensive discussion of long-short optimization, and newtools for dealing with estimation error in applications and enhancing computationalefficiency. RE optimization is shown to be a Bayesian-basedgeneralization and enhancement of Markowitz's solution. RE technologycorrects many current practices that may adversely impact theinvestment value of trillions of dollars under current asset management. RE optimization technology may also be usefulin other financial optimizations and more generally in multivariateestimation contexts of information uncertainty with Bayesian linearconstraints. Summary: Important information when considering Markowitz optimization Rating: 4 Michaud's resampling methodology is quite rigorous, although thepatentability of application of econometric concepts that are over 40years old to a theory advanced by Markowitz in 1952 should be seriouslyquestioned by any rational reader. The applicability of resampling andimprovements to the inputs to estimation are clear, and should bestrongly considered by anyone in the asset management industry. Thebook glosses over other approaches to optimization that are not basedon Markowitz, all but ignoring the huge body of literature that hasbeen built up around other optimization approaches. Except for thisshortfall, this is an excellent book, and shoulds be a part of your library on quantitative asset management. 1 Introduction 3 Markowitz Effi ciency 3 An Asset Management Tool 4 Traditional Objections 5 The Most Important Limitations 5 Resolving the Limitations of Mean-Variance Optimization 6 Illustrating the Techniques 6 2 Classic Mean-Variance Optimization 7 Portfolio Risk and Return 7 Defi ning Markowitz Effi ciency 9 Optimization Constraints 9 The Residual Risk-Return Effi cient Frontier 10 Computer Algorithms 10 Asset Allocation Versus Equity Portfolio Optimization 11 A Global Asset Allocation Example 13 Reference Portfolios and Portfolio Analysis 14 Return Premium Effi cient Frontiers 16 Appendix: Mathematical Formulation of MV Effi ciency 17 3 Traditional Criticisms and Alternatives 20 Alternative Measures of Risk 20 Utility Function Optimization 22 Multiperiod Investment Horizons 23 Asset-Liability Financial Planning Studies 25 Linear Programming Optimization 27 4 Unbounded MV Portfolio Effi ciency 29 Unbounded MV Optimization 30 The Fundamental Limitations of Unbounded MV Effi ciency 31 Repeating Jobson and Korkie 32 Implications of Jobson and Korkie Analysis 33 Statistical MV Effi ciency and Implications 34 5 Linear Constrained MV Effi ciency 35 Linear Constraints 35 Effi cient Frontier Variance 37 Rank-Associated Effi cient Portfolios 39 How Practical an Investment Tool? 40 6 The Resampled Effi cient Frontier™ 42 Effi cient Frontier Statistical Analysis 42 Properties of Resampled Effi cient Frontier Portfolios 45 True and Estimated Optimization Inputs 47 Simulation Proofs of Resampled Effi ciency Optimization 48 Why Does It Work 51 Certainty Level and RE Optimality 51 FC Level Applications 52 The REF Maximum Return Point (MRP) 53 Implications for Asset Management 55 Conclusion 55 Appendix A: Rank- Versus λ-Associated RE Portfolios 56 Appendix B: Robert’s Hedgehog 57 7 Portfolio Rebalancing, Analysis, and Monitoring 60 Resampled Effi ciency and Distance Functions 61 Portfolio Need-to-Trade Probability 62 Meta-Resampling Portfolio Rebalancing 63 Portfolio Monitoring and Analysis 64 Conclusion 66 Contents xv Appendix: Confi dence Region for the Sample Mean Vector 66 8 Input Estimation and Stein Estimators 68 Admissible Estimators 69 Bayesian Procedures and Priors 69 Four Stein Estimators 70 James-Stein Estimator 70 James-Stein MV Effi ciency 71 Out-of-Sample James-Stein Estimation 72 Frost-Savarino Estimator 73 Covariance Estimation 74 Stein Covariance Estimation 76 Utility Functions and Input Estimation 77 Ad Hoc Estimators 77 Stein Estimation Caveats 78 Conclusions 78 Appendix: Ledoit Covariance Estimation 78 9 Benchmark Mean-Variance Optimization 80 Benchmark-Relative Optimization Characteristics 80 Tracking Error Optimization and Constraints 81 Constraint Alternatives 83 Roll’s Analysis 85 Index Effi ciency 85 A Simple Benchmark-Relative Framework 86 Long-Short Investing 86 Conclusion 88 10 Investment Policy and Economic Liabilities 89 Misusing Optimization 90 Economic Liability Models 90 Endowment Fund Investment Policy 91 Pension Liabilities and Benchmark Optimization 92 Limitations of Actuarial Liability Estimation 92 Current Pension Liabilities 93 Total and Variable Pension Liabilities 93 Economic Signifi cance of Variable Liabilities 94 Economic Characteristics of VBO Liabilities 95 xvi Contents An Example: Economic Liability Pension Investment Policy 96 Past and Future of Defi ned Benefi t Pension Plans 98 Conclusion 99 11 Bayes and Active Return Estimation 101 Current Practices 102 Bayes Principles 102 The Bayes Return Formula 102 A Bayes Panel Illustration 103 Bayesian Mixed Estimation Issues 104 Enhanced Inputs or Enhanced Optimizer 106 Bayesian Caveats 107 12 Avoiding Optimization Errors 109 Scaling Inputs 109 Financial Reality 111 Liquidity Factors 111 Practical Constraint Issues 112 Biased Portfolio Characteristics 112 Index Funds and Optimizers 113 Optimization from Cash 114 Forecast Return Limitations 115 Conclusion 116 Epilogue 117 Bibliography 119 Index 125 |
[biggrin]
京公网安备 11010802022788号
