Investor beliefs and forecast evaluation [Elektronische Ressource] / Qingwei Wang

Informations

• Publié par: universitat_mannheim
• Publié le: 01 janvier 2010
• Nombre de lectures: 18
• Langue: English

Extrait

Qingwei Wang
Investor Beliefs and Forecast
Evaluation
Inaugural-Dissertation
zur Erlangung des akademischen Grades
eines Doktors der Wirtschaftswissenschaften
der Universität Mannheim
Vorgelegt im Sommersemester 2010Dekan: Prof. Dr. Hans H. Bauer
Referent: Prof. Dr. Ernst Maug
Korreferent: Prof. Dr. Stefan Ruenzi
Tag der mündlichen Prüfung: 03. März 2010ACKNOWLEDGEMENTS
Writing this dissertation is an exceptional journey with both sweets and joys. Yet it
won’t be possible to get finished without support of numerous individuals.
At the outset, I would like to thank my supervisor, Prof. Ernst Maug, for his support
and patience in the past four years. I am fortunate to have the opportunity to write
the dissertation under his supervision. He has constantly encouraged me to remain
focused on achieving my goal, inspiring and guiding me to pursue an academic career
in finance. I appreciate his thorough and enlightening comments on my papers.
I am also grateful to Prof. Stefan Ruenzi, a member of my dissertation committee. He
was very diligent at commenting my article, suggesting new approaches and guiding my
research direction. I benefit tremendously from his encouragement and illuminating
advice.
My thanks go also to Dr. Michael Schröder. During the course of my stay as a re-
searcher at the Centre for European Economic Research, I have benefited greatly from
interactions with him, both from joint research projects and from the joint research
paper. I am indebted to him for providing me the freedom for conducting my research.
Many local “peers” at the Centre for European Economic Research and University
of Mannheim have contributed towards shaping this thesis. I am indebted to Prof.
Francois Laisney, Waldemar Rotfuss, Sandra Schmidt for their critical comments
and helpful suggestions. I also thank my coauthors, Chritian Dick, Prof. Bernd
Fitzenberger, Karsten Kohn, Pei Kuang, Christoph Schneider, Andreas Schrimpf and
Michael Schröder, whose insights and interactions keep the joint research a particularly
enjoyable experience.
iMany thanks to Hela Hellerich, Richard Hinz, Nataliya Matosova, Frieder Mokinski
and Jonas Vogt for their excellent research assistance. Hela Hellerich has provided
careful proof-reading of the different chapters of this thesis.
Finally, I wish to extend my special thanks to my parents, Fangying Li and Dezhi Wang,
and my sister, Yun Wang, for their love and encouragement in the past years. I also
thank my parents in-laws for their invaluable support. Of course, I am immensely
grateful to my wife, Ru Xie, who provided a constant source of love, comfort, and
stability through the ups and downs during the course of this thesis.
Qingwei Wang
iiCONTENTS
Acknowledgements i
Preface 1
1 Sentiment, Convergence of Opinion, and Market Crash 9
1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.2 Literature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1.3 Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
1.4 Empirical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
1.4.1 Skewness in Different Quintiles of Sentiment and Disagreement 22
1.4.2 Regression Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
1.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Appendix: Trading Rules Documentation . . . . . . . . . . . . . . . . . . . . . 41
2 How Illusory is the Profitability of Technical Analysis? 44
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
2.2 Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
iiiCONTENTS
2.3 Universe of Trading Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
2.4 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
2.4.1 Performance Measures . . . . . . . . . . . . . . . . . . . . . . . . . 56
2.4.2 Empirical Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
2.5 Empirical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
2.5.1 Full Sample Results . . . . . . . . . . . . . . . . . . . . . . . . . . 60
2.5.2 Sub-sample Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 67
2.5.3 Out-of-sample Data Snooping Bias . . . . . . . . . . . . . . . . . . 71
2.5.4 Reduced Universe of Trading Rules . . . . . . . . . . . . . . . . . 74
2.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
Appendix A: Description of Simple Trading Rules . . . . . . . . . . . . . . . . 77
Appendix B: Documentation of Trading Rules Parameters . . . . . . . . . . . 79
Appendix C: Tests without Data Snooping Bias . . . . . . . . . . . . . . . . . . 84
3 A Reappraisal of the Leading Indicator Properties of the Yield Curve
in the Presence of Structural Instability 89
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
3.2 The Predictive Power of the Yield Spread: A Reexamination . . . . . . . 93
3.3 Empirical Analysis of Model Instability and Forecast Breakdowns . . . . 103
3.4 The Role of Other Yield Curve Information . . . . . . . . . . . . . . . . . 113
3.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
ivCONTENTS
Appendix A: Data Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
Appendix B: Estimating Return Risk Premia . . . . . . . . . . . . . . . . . . . 121
Bibliography 123
vLIST OF TABLES
1.1 Summary Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
1.2 Correlation Coefficient Matrix . . . . . . . . . . . . . . . . . . . . . . . . . 18
1.3 Correlation with Other Sentiment Indicators . . . . . . . . . . . . . . . . 21
1.4 Breakdown by Sorts of Sentiment and Disagreement . . . . . . . . . . . 23
1.5 Breakdown by Sorts of Disagreement in Different States of Sentiment . 24
1.6 Unit Root Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
1.7 Forecasting the Aggregate Stock Market Crash (DJIA 1952-2008) . . . . 29
1.8 Forecasting the Aggregate Stock Market Crash (DJIA 1900-1951) . . . . 32
1.9 Forecasting the Aggregate Stock Market Crash (S&P 500 Index 1964-2008) 35
1.10 Forecasting the Aggregate Stock Market Crash at Monthly Frequency
(DJIA 1952-2008) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
1.11 Forecasting the Aggregate Stock Market Crash with Learning (DJIA
1952-2008) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
1.12 Forecasting the Aggregate Stock Market Crash with Learning (S&P 500
Index 1964-2008) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
2.1 The Universe of Trading Rules . . . . . . . . . . . . . . . . . . . . . . . . 51
viLIST OF TABLES
2.2 Data Used for Testing the Profitability of Trading Rules . . . . . . . . . 60
2.3 Summary Statistics for Daily Changes in the Logarithm of Exchange Rates 60
2.4 Performance of the Best FX Trading Rules in Emerging Market (without
Transaction Cost) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
2.5 Performance of the Best FX Trading Rules in Emerging Market (with a
0.1% one-way Transaction Cost) . . . . . . . . . . . . . . . . . . . . . . . . 65
2.6 Number of Profitable Trading Rules . . . . . . . . . . . . . . . . . . . . . 68
2.7 Performance of the Best FX Trading Rules in First Sub-period (with a
0.1% one-way Transaction Cost) . . . . . . . . . . . . . . . . . . . . . . . . 69
2.8 Performance of the Best FX Trading Rules in Second Sub-period (with a
0.1% one-way Transaction Cost) . . . . . . . . . . . . . . . . . . . . . . . . 70
2.9 Number of Profitable Trading Rules in Out-of-Sample Experiment . . . 72
2.10 Number of Profitable Trading Rules (Reduced Universe) . . . . . . . . . 75
3.1 Predictive Regressions for Real GDP Growth using the Yield Spread . . 96
3.2 Out-of-Sample Performance of the Yield Spread: Forecast Evaluation
Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
3.3 Structural Break Tests: Predictive Regressions for Real GDP Growth . . 108
3.4 Identification of Break Dates . . . . . . . . . . . . . . . . . . . . . . . . . 109
3.5 Window Selection under Model Instability: Forecasting Evaluation Statis-
tics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
3.6 Predictive Content of the Term Spread and other Yield Curve Variables 115
vii