Harry Markowitz

The following sections are included:

• Foreword

• Contents

• Acknowledgements

The following sections are included:

• TRAINS OF THOUGHT
  • What Do We Know?
  • Probability, Utility and Quadratic Approximations
  • Simulation and Systems Descriptions
  • Personal Reflections
  • Notes
  • References

The following sections are included:

• PORTFOLIO SELECTION*

• The Early History of Portfolio Theory: 1600–1960
  • Portfolio Theory: 1952
  • Markowitz Portfolio Theory circa 1959
  • Tobin (1958)
  • Hicks (1935,1962)
  • Marschak (1938)
  • Williams (1938)
  • Leavens (1945)
  • The End of the Beginning
  • Notes
  • References

• THE UTILITY OF WEALTH

The following sections are included:

• INDUSTRY-WIDE, MULTI-INDUSTRY AND ECONOMY-WIDE PROCESS ANALYSIS
  • Introduction
  • Subanalysis for the petroleum refining industry
  • Problems of economy-wide analysis
  • Process analysis and other models of the economy
  • A list of works cited

• ALTERNATE METHODS OF ANALYSIS
  • Introduction
  • Gross National Product Analysis
  • Requirements Analysis
  • Input-Output Analysis
  • Summary
  • References

• THE ELIMINATION FORM OF THE INVERSE AND ITS APPLICATION TO LINEAR PROGRAMMING
  • Introduction
  • The Elimination Form of Inverse
  • Application to Linear Programming

• THE OPTIMIZATION OF A QUADRATIC FUNCTION SUBJECT TO LINEAR CONSTRAINTS
  • QUADRATIC PROBLEMS
  • ASSUMPTIONS
  • THE CRITICAL LINE
  • CRITICAL LINES ℓ(ℓ,ℓ)
  • INTERSECTIONS OF CRITICAL LINES; NON-DEGENERACY CONDITIONS
  • THE ALGORITHM UNDER CONDITIONS 1 THROUGH 4
  • THE ALGORITHM UNDER CONDITIONS 3 AND 4
  • THE ALGORITHM WHEN L_E IS DEGENERATE BUT UNIQUE
  • THE ALGORITHM WHEN L_E IS NOT UNIQUE
  • THE ALGORITHM, WHEN CONDITION 3 DOES NOT HOLD
  • THE SET OF EFFICIENT E, V COMBINATIONS
  • MINIMIZING A QUADRATIC

• The general mean-variance portfolio selection problem
  • The problem
  • Application
  • Computation
  • Degeneracy and other problems
  • References

The following sections are included:

• SIMULATING WITH SIMSCRIPT
  • Simscript's World-View
  • The Simscript Method of Communication
  • References

• PROGRAMMING BY QUESTIONNAIRE
  • PREFACE
  • CONTENTS
  • INTRODUCTION
  • THE MECHANICS OF PROGRAM GENERATION
    • USING A PROGRAM GENERATOR
    • THE QUESTIONNAIRE
    • THE STATEMENT LIST
    • DECISION TABLES
    • THE EDITOR
  • OBSERVATIONS AND DISCUSSION
    • OTHER TECHNIQUES
    • EXTENSIONS OF THE PROGRAM GENERATION CONCEPT
    • CHANGING THE GENERATOR AND MODIFYING GENERATED PROGRAMS
    • VARIANTS OF THE QUESTIONNAIRE
    • DIFFICULTIES YET UNSOLVED
    • CONCLUSION
  • Appendix A QUESTIONNAIRE
  • Appendix B SAMPLE STATEMENT LIST AND DECISION TABLES

• SIMSCRIPT
  • HISTORY
  • BASIC CONCEPTS
    • Entities
    • Attributes
    • Sets
    • Change of Status
  • SIMULATION APPLICATIONS
    • A Job Shop Exercise
    • A Life Insurance Company Corporate Model
    • A VM/370 Simulator
  • AN ENTITY, ATTRIBUTE, SET, AND EVENT VIEW OF DATA BASE SYSTEMS
    • Comparison with Other Data Base Languages
  • THE SIMSCRIPT LANGUAGE WRITING LANGUAGE
    • Additional Level 7 Needs
  • ENTITIES OF LOGIC AND MATHEMATICS
  • SUMMARY
  • REFERENCES

• BARRIERS TO THE PRACTICAL USE OF SIMULATION ANALYSIS
  • INTRODUCTION
  • PAST
  • PRESENT
  • FUTURE
  • POSTSCRIPT AND SUMMARY
  • REFERENCES

The following sections are included:

• Approximating Expected Utility by a Function of Mean and Variance
  • A Class of Approximations
  • Analysis of Error Functions
  • Empirical Results
  • Some Objections Reconsidered
  • The E,V Investor
  • REFERENCES

• Mean-variance Versus Direct Utility Maximization
  • The Problem
  • The Quality of the Approximation
  • The Selected Utility Functions
  • The Data
  • The Empirical Results
  • The Effect of Leverage
  • Conclusions
  • Appendix The Direct Maximization Algorithm
    • The Problem
    • Details on the Maximization Method of Step 2
  • REFERENCES

• The Value of a Blank Check
  • CONCEPTS OF A BLANK CHECK LOTTERY
  • A SURVEY
  • A PROBLEM WITH EXPONENTIAL UTILITY
  • THE EFFECTS OF HUMAN CAPITAL
  • APPLICATION TO EVALUATING MEAN-VARIANCE APPROXIMATIONS
  • SUMMARY AND CONCLUSION
  • APPENDIX
  • ENDNOTES
  • REFERENCES

• The “two beta” trap
  • NORMATIVE AND POSITIVE MEAN-VARIANCE ANALYSIS
  • PROPOSITIONS ABOUT BETAS
  • EPILOGUE AND EDITORIAL
  • REFERENCES

• Portfolio Analysis with Factors and Scenarios
  • The Model
    • The Usual States of the World Model
    • Modeling the ɛ's as being Conditionally Uncorrelated
    • Combining Scenarios and Factors
  • Diagonalization
  • Conclusion
  • REFERENCES

• SPARSITY AND PIECEWISE LINEARITY IN LARGE PORTFOLIO OPTIMIZATION PROBLEMS
  • INTRODUCTION
  • MODELS FOR THE ESTIMATION OF COVARIANCES
  • EXISTING OPTIMIZATION METHODS
  • REVIEW OF THE COMPLEMENTARY PIVOT ALGORITHM
    • Getting started
    • Basis changes
  • EXPLOITING THE STRUCTURE OF C
    • Use with a sparse matrix package
    • Implementation out-of-core
  • IMPLICIT TREATMENT OF UPPER AND LOWER BOUNDS AND TRANSACTIONS COSTS
  • CONCLUSIONS AND SUMMARY
  • REFERENCES
  • DISCUSSION

• The ER and EAS Formalisms for System Modeling and the EAS–E Language
  • INTRODUCTION
  • THE EAS WORLDVIEW
  • EAS and ER
  • THE EAS-E LANGUAGE
  • THE REST OF THE ICEBERG
  • THE EAS PRINCIPLE
  • APPLICABILITY TO ER
  • REFERENCES

• EAS-E: An Integrated Approach to Application Development
  • INTRODUCTION
  • THE EAS MODEL
  • THE EAS-E PROGRAMMING LANGUAGE
    • Sets as Standard Data Structures
    • Integrated Language
    • Some Examples of EAS-E Syntax
      • The FOR EACH Phrase
      • The FIND Statement
      • Full-Screen Input/Output Capability
    • Entity, Attribute, and Set Definitions
    • A PL-I/SQL Program and an Equivalent EAS-E Program
    • A PLAIN Program and an Equivalent EAS-E Program
  • THE BROWSE FACILITY
  • SOME IMPLEMENTATION DETAILS
    • Database Ranked Sets
    • Reference and Identifier Variables
    • Recording and Unlocking
  • THE EAS-E DATABASE MANAGEMENT SYSTEM
    • Concurrency Control
    • Protecting Against Software Crashes
    • Protecting Against Physical Damage
  • SUMMARY AND STATUS
  • REFERENCES

• The System Architecture of EAS-E: An Integrated Programming and Database Language
  • Very little is required to go from the conceptual model to the application program
  • To write a program that works with the data base, the user must specify which entity types are to be maniplated
  • Simple queries can be written as small EAS-E programs
  • The problem of passing the selection information at execution time to the loop-searching mechanism has been addressed by designing an EAS structure to contain that information
  • EAS-E has been designed to accommodate data bases of arbitrary size, from very small to very large

• Samuelson and Investment for the Long Run
  • Background
  • The Expected Log Rule in General and Particular
  • First Argument For Max E log
  • Argument Against Max E log
  • Example
  • Another Argument For Max E log
  • Summary
  • References

• INVESTMENT FOR THE LONG RUN: NEW EVIDENCE FOR AN OLD RULE
  • Background
  • The Sequence of Games
  • Alternate Sequence-of-Games Formalizations
  • Unending Games
  • Conclusions
  • Appendix
  • References

The following sections are included:

• Investment Rules, Margin and Market Volatility
  • THE SIMULATED MARKET
  • EFFECT OF VARYING THE NUMBER OF PORTFOLIO INSURERS
  • CONCLUSIONS, CAVEATS, AND CONJECTURES
  • APPENDIX
  • REFERENCES

• Risk Adjustment
  • Traditional CAPMs
  • Risk Adjustment in the Standard, Homogeneous Model
  • Observations and Extensions
  • Epilogue
  • REFERENCES

• Normative Portfolio Analysis: Past, Present and Future
  • Normative Portfolio Analysis as of 1959
  • Normative versus Positive Portfolio Analysis
  • Progress and Opportunity in Normative Analysis
  • References

• Individual versus Institutional Investing
  • Thesis
  • Antithesis
  • Synthesis
  • Notes
  • References

• Foundations of Portfolio Theory

• Fast Computation of Mean–variance Efficient Sets Using Historical Covariances
  • INTRODUCTION
  • REFORMULATION OF THE PROBLEM
  • IMPLEMENTATION OF THE CRITICAL LINE ALGORITHM
  • PERFORMANCE
  • APPENDIX
  • NOTES
  • REFERENCES

• Computation of mean-semivariance efficient Sets by the Critical Line Algorithm
  • Introduction
  • Review of mean-variance model
    • THE MEAN-VARIANCE PROBLEM
    • THE CRITICAL LINE ALGORITHM
  • Mean-semivariance model
    • THE MEAN·SEMIVARIANCE PROBLEM
    • REFORMULATION OF THE PROBLEM
  • Implementation of the Critical Line Algorithm
  • Performance
  • References

• Data Mining Corrections
  • THE MODELS
  • ESTIMATION OF β FOR MODEL I
  • ESTIMATION OF β FOR MODEL II
  • TESTS OF SIGNIFICANCE
  • ESTIMATION FOR MODEL III
  • A BAYESIAN VIEW OF THE METHODS
  • EXPERIENCE WITH MODELS I, II, AND III
  • WHY NO HOLDOUT PERIOD?
  • SUMMARY
  • ENDNOTES
  • REFERENCES

The following sections are included:

• The Likelihood of Various Stock Market Return Distributions: Part 1: Principles of Inference
  • Financial research supports financial decision making: An example
  • Rational (coherent) decision making is Bayesian
  • Classical statistics is an unreliable indicator of how Bayesians should shift beliefs
  • Remote Bayesian clients
  • Human approximation to an RDM
  • Summary
  • Acknowledgments
  • Notes
  • References

• The Likelihood of Various Stock Market Return Distributions, Part 2: Empirical Results
  • Simple hypotheses
    • Hypotheses
    • The sample
    • Computation of likelihood
    • Shifts in beliefs
  • Compound hypotheses
    • Is Y_t Gaussian?
    • Is Y_t generated by a student's t distribution?
    • Is Y_t contaminated normal?
    • On estimating ψ
  • What next?
  • Summary and conclusions
  • Acknowledgments
  • Notes
  • References

• RESAMPLED FRONTIERS VERSUS DIFFUSE BAYES: AN EXPERIMENT
  • Introduction
  • The Michaud Player
  • The diffuse Bayes player
    • Basics
    • Diffuse priors
    • Importance sampling
  • Results
  • Questions
  • Conclusions
  • Acknowledgments
  • Notes
  • References

• On Socks, Ties and Extended Outcomes
  • Extensions and Reflections
  • Appendix Axioms for Multiperiod Analysis
  • References
  • Endnotes

• Single-Period Mean–Variance Analysis in a Changing World
  • The Model
    • Computation of Expected Discounted Utility
    • Computation of (Nearly) Optimal Strategy
    • (Nearly) Optimum Action Matrixes
  • MV Heuristic
    • Comparison of Expected Utilities
    • Where Next?
  • Conclusion
  • Appendix A. Why A* May Not Equal Ã

• Financial Market Simulation
  • TYPES OF DYNAMIC MODELS
  • JLM SIMULATOR
    • Status
    • Events
  • OBJECTIVES AND EXTENSIONS
    • Alternative Investor and Trader Behaviors
    • Model Size
  • ADVANTAGES OF ASYNCHRONOUS FINANCE MODELS
  • CAVEAT
  • CONCLUSION
  • ENDNOTES
  • REFERENCES

• Portfolio Optimization with Factors, Scenarios, and Realistic Short Positions
  • Introduction
  • The General Mean-Variance Problem
  • Solution to the General Problem
  • Diagonizable Models of Covariance
  • Short Sales in Practice
  • Modeling Short Sales
  • Solution to Long-Short Model
  • Example
  • Summary
  • Endnotes
  • References

• Market Efficiency: A Theoretical Distinction and So What?
  • A Distinction
  • Generalizations
  • So What?
  • Conclusion
  • Appendix A. Finding a Probability Distribution for a Given Efficient Set
  • Notes
  • References

• EFFICIENT PORTFOLIOS, SPARSE MATRICES, AND ENTITIES: A RETROSPECTIVE
  • PORTFOLIO THEORY
  • SPARSE MATRICES
  • ENTITIES, ATTRIBUTES, SETS, AND EVENTS
  • ENDNOTE
  • REFERENCES

• De Finetti Scoops Markowitz
  • The News
  • The De Finetti Model
  • Numerical Example
  • Summary
  • Technical Supplement
    • Solution to the de Finetti problem
    • A Correct Final Segment Theorem
    • Advances in Mathematical Programming (1940-1956)
  • References

• CAPM Investors Do Not Get Paid for Bearing Risk: A Linear Relation Does Not Imply Payment for Risk
  • THE MOSSIN VERSION OF THE SHARPE-LINTNER MODEL
  • AFTERTHOUGHTS
  • WHERE DOES THIS LEAVE THE CAPM?
  • ENDNOTE
  • REFERENCES