Developments in Forecast Combination and

Developments in Forecast Combination and Portfolio Choice focuses on the following three themes: model and forecast combinations; structural change and long memory, controlling downside risk and investment strategies. Written by leading international researchers and practitioners, his book deals efficiently with three key questions facing portfolio managers. How to achieve greater forecasting accuracy; how to deal with structural change in asset allocation models and how to control downside risk, i.e. the risk of loss, in portfolio management.

CHRISTIAN L. DUNIS is Girobank Professor of Banking and Finance at Liverpool Business School, and Director of its Centre for International Banking, Economics and Finance (CIBEF). He is also a consultant to asset management firms and an Official Reviewer attached to the European Commission for the evaluation of applications to Finance of emerging software technologies. He is an Editor of the European Journal of Finance and has published widely in the field of financial market analysis and forecasting. ALLAN TIMMERMANN is Professor of Economics at University of California, San Diego. He is on the editorial board of the Journal of Forecasting and Journal of Business and Economic Statistics. His research is concerned with modelling the dynamics and predictability of returns in financial markets. Professor Timmermann has held positions at Birkbeck College and the London School of Economics. JOHN MOODY is the Director of the Computational Finance program and a Professor of Computer Science at the Oregon Graduate Institute. His research interests include computational finance, time series analysis and machine learning. Professor Moody has held positions at Yale University and the Institute for Theoretical Physics.

Contributors

ix

About the Contributors

xi

Series Preface

xvii

Preface

xxi

THEME I MODEL AND FORECAST COMBINATIONS

1

(142)

What Exactly Should We Be Optimising? Criterion Risk in Multicomponent and Multimodel Forecasting

3

(24)

A. Neil Burgess

Abstract

3

(1)

Introduction

3

(2)

Model Combination and Criterion Risk

5

(5)

A Population-based Algorithm to Perform Joint Optimisation of a Portfolio of Models

10

(4)

Simulation Results with Synthetic Data

14

(6)

Empirical Results for a Portfolio of Statistical Arbitrage Models

20

(5)

Conclusion

25

(2)

References

26

(1)

A Meta-parameter Approach to the Construction of Forecasting Models for Trading Systems

27

(18)

Neville Towers

A. Neil Burgess

Abstract

27

(1)

Introduction

27

(2)

Developing Forecasting Models for Trading Systems

29

(3)

Synthetic Example of the Joint Optimisation Technique

32

(6)

Application to Statistical Arbitrage Trading

38

(5)

Conclusions

43

(2)

References

44

(1)

The Use of Market Data and Model Combination to Improve Forecast Accuracy

45

(36)

Christian L. Dunis

Jason Laws

Stephane Chauvin

Abstract

45

(1)

Introduction

45

(1)

The Exchange Rate and Volatility Data

46

(5)

The Volatility Models and Estimation Procedure

51

(6)

The Out-of-Sample Estimation Results

57

(12)

Conclusion

69

(12)

References

70

(2)

Appendix 3.1

72

(3)

Appendix 3.2

75

(3)

Appendix 3.3

78

(1)

Appendix 3.4

78

(3)

21 Nonlinear Ways to Beat the Market

81

(36)

George T. Albanis

Roy A. Batchelor

Abstract

81

(1)

Introduction

81

(1)

Data and Trading Rules

82

(4)

Classification Methods

86

(10)

Implementation

96

(4)

Results

100

(7)

Conclusion

107

(10)

Acknowledgements

108

(1)

References

108

(2)

Appendix 4.1

110

(7)

Predicting High Performance Stocks Using Dimensionality Reduction Techniques Based on Neural Networks

117

(26)

George T. Albanis

Roy A. Batchelor

Abstract

117

(1)

Introduction

117

(1)

Principal Components Analysis

118

(2)

Neural Network Linear Principal Components Analysis

120

(1)

Neural Network Nonlinear Principal Components Analysis

121

(3)

Data and Forecast Methodology

124

(3)

Results

127

(6)

Conclusion

133

(10)

Acknowledgements

133

(1)

References

133

(1)

Appendix 5.1

134

(9)

THEME II STRUCTURAL CHANGE AND LONG MEMORY

143

(54)

Structural Change and Long Memory in Volatility: New Evidence from Daily Exchange Rates

145

(14)

Michael Beine

Sebastien Laurent

Abstract

145

(1)

Introduction

145

(1)

Preliminary Evidence

146

(2)

The Single-regime FIGRACH Model

148

(1)

The Markov-Switching FIGARCH Model

149

(7)

Conclusion

156

(3)

Acknowledgements

156

(1)

References

156

(3)

Long-run Volatility Dependencies in Intraday Data and Mixture of Normal Distributions

159

(20)

Aurelie Boubel

Sebastien Laurent

Abstract

159

(1)

Introduction

159

(1)

The Intraday Volatility

160

(9)

Long Memory from Intraday Returns

169

(5)

Conclusion

174

(5)

Acknowledgements

175

(1)

References

175

(2)

Appendix 7.1: Data Construction

177

(1)

Appendix 7.2: Data Transformation

177

(2)

Comparison of Parameter Estimation Methods in Cyclical Long Memory Time Series

179

(18)

Laurent Ferrara

Dominique Guegan

Abstract

179

(1)

Introduction

179

(3)

Semiparametric Estimation

182

(4)

Pseudo-Maximum Likelihood Estimation

186

(2)

Application

188

(5)

Conclusion

193

(4)

Acknowledgements

194

(1)

References

194

(3)

THEME III CONTROLLING DOWNSIDE RISK AND INVESTMENT STRATEGIES

197

(102)

Building a Mean Downside Risk Portfolio Frontier

199

(14)

Gustavo M. de Athayde

Abstract

199

(1)

Introduction

199

(2)

The Mean DSR Portfolio Frontier: The Bivariate Case

201

(4)

The Algorithm

205

(2)

The Multivariate Case

207

(2)

Asset Pricing

209

(2)

Conclusion

211

(2)

Acknowledgements

211

(1)

References

211

(2)

Implementing Discrete-time Dynamic Investment Strategies with Downside Risk: A Comparison of Returns and Investment Policies

213

(18)

Mattias Persson

Abstract

213

(1)

Introduction

213

(1)

Investment Models

214

(4)

Data

218

(1)

Results

219

(9)

Summary and Concluding Remarks

228

(3)

References

228

(3)

Portfolio Optimisation in a Downside Risk Framework

231

(8)

Riccardo Bramante

Barbara Cazzaniga

Abstract

231

(1)

Introduction

231

(1)

Methodology

232

(2)

Empirical Evidence

234

(2)

Summary and Concluding Remarks

236

(3)

References and Bibliography

236

(3)

The Three-moment CAPM: Theoretical Foundations and an Asset Pricing Model Comparison in a Unified Framework

239

(36)

Emmanuel Jurczenko

Bertrand Maillet

Abstract

239

(1)

Introduction

239

(3)

Investor's Preferences and the Three-moment CAPM

242

(5)

The Three-moment CAPM

247

(15)

Conclusion

262

(13)

Acknowledgements

262

(1)

References

263

(3)

Appendix 12.1

266

(1)

Appendix 12.2

266

(1)

Appendix 12.3

267

(1)

Appendix 12.4

268

(1)

Appendix 12.5

269

(1)

Appendix 12.6

269

(1)

Appendix 12.7

270

(1)

Appendix 12.8

271

(1)

Appendix 12.9

271

(1)

Appendix 12.10

272

(3)

Stress-testing Correlations: An Application to Portfolio Risk Management

275

(24)

Frederick Bourgoin

Abstract

275

(1)

The Fall of LTCM and the Credit Crisis of August 1998

275

(5)

Stressing the General Level of Correlation of a Given Portfolio

280

(9)

Stress-testing Correlation Matrices

289

(8)

Conclusion

297

(2)

References

297

(2)

Index

299

What is included with this book?

The New copy of this book will include any supplemental materials advertised. Please check the title of the book to determine if it should include any access cards, study guides, lab manuals, CDs, etc.

The Used, Rental and eBook copies of this book are not guaranteed to include any supplemental materials. Typically, only the book itself is included. This is true even if the title states it includes any access cards, study guides, lab manuals, CDs, etc.

FREE SHIPPING

Amazon no longer offers textbook rentals. We do!

Developments in Forecast Combination and Portfolio Choice

Summary

Author Biography

Table of Contents

Supplemental Materials