Three Essays in Financial Economics Open Access

Noh, Joonki (2015)

Permanent URL:


This dissertation covers the information dissemination in financial markets and the risk-return relationship in the cross-section of stocks. The first essay explores the information diffusion to equity markets in the framework of networks. I investigate whether the number of connections that an industry has in the network of inter-industry trade affects the speed of information flow to the industry. I find that the information flows more slowly to central industries from their related (=customer and supplier) industries than to peripheral industries from their related industries. The strong return predictability to central industries from related industries leads to highly profitable trading strategies whose risk-adjusted returns are 7.0% to 7.9% per annum. To explain this finding, I argue that investors who invest in central industries need to process more complicated information about related industries, slowing down the information flow to central industries. I find evidence that the sell-side analysts of central industries also face more complicated information about related industries, slowing down their processing new information about related industries. The network framework helps to identify an unknown and unique anomaly and to better understand potential sources of anomalies. The second essay develops an IV methodology to test asset pricing models using individual stocks as test assets. We obtain consistent estimates of risk premiums, and simulation indicates that the associated tests are well specified in small samples. When testing three asset pricing models known to be successful in the literature when they were tested with characteristics-sorted portfolios as test assets, we find weak evidence that their factor risks are reliably priced in the cross-section. The third essay investigates whether the facial expressions in CEOs' televised interviews can convey information about firms to investors and whether investors understand and react to it. We find evidence that negative facial expressions are correlated with CARs and turnover over the next one to two days after air dates. We also find that negative facial expressions are associated with firms' one-quarter-ahead earnings. Taken together, it presents the first evidence in financial economics that CEOs' facial expressions in their televised interviews can be a channel through which value-relevant information is disseminated.

Table of Contents

Industry Networks and the Speed of Information Flow 1

Introduction 2

Related Literature 7

Methodology, Data, and Preliminary Analyses 9

Methodology for the Industry Network 9

Other Data Sources and Variables 12

Preliminary Analyses 14

Empirical Results 16

Return Cross-predictability 17

Quantifying the Economic Magnitudes 19

Potential Economic Mechanism 21

Distinction from Existing Anomalies 25

Change in Institutional Co-ownership 29

Conclusions 30

Technical Appendix 32

Details on Constructing the Industry Network 32

Empirical Tests of Asset Pricing Models with Individual Stocks 33

Introduction 34

Methodology 36

Small Sample Properties 42

Return Generating Processes 42

Simulation Experiments: Parameters and Methodology 43

EIV Bias 45

Small Sample Distribution of the Test Statistic 47

Empirical Results 47

Data 47

CAPM and Fama-French Three-factor Model 48

Production-based Asset Pricing Model 52

Liquidity-adjusted CAPM 55

Strength of Instruments 59

Conclusion 61

Technical Appendix 63

Proofs of Propositions 63

Simulation Parameters 73

Innovations in Illiquidity Costs 74

Proof of Proposition 3 75

Information in CEO's Facial Expressions: A First Look 76

Introduction 77

Quantifying Emotions based on Facial Expressions 79

Video Sample Selection 81

Empirical Results 82

Preliminary Results 82

Facial Expressions and Cumulative Abnormal Returns 85

Facial Expressions and Cumulative Abnormal Turnover 88

Facial Expressions and Future Earnings 90

Discussions 91

Conclusion 92

Technical Appendix 93

Variable Definition 93

Links to Sample Videos 95

Screenshots from the Videos with the Most Extreme Facial Expression Scores 96

Appendices 99

Tables 99

Figures 133

References 140

About this Dissertation

Rights statement
  • Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
  • English
Research Field
Committee Chair / Thesis Advisor
Committee Members
Last modified

Primary PDF

Supplemental Files