Ebook Information Structure And Stock Return Distribution; An Empirical Examination
Ever since the introduction of the theory of portfolio selection in Markowitz (1952) and Tobin (1958), many financial economists and statisticians have been concerned with the description of stock returns.
The specification of stock return distributions has had a significant impact on the asset pricing models developed in the finance literature. For example, the normality assumption of the stock return distributions is crucial to the development of the mean-variance portfolio theory, and an understanding of the behavior of stock return variance is essential to the option pricing models.
The most convenient assumption for financial theory and statistical testing is that the distribution of stock returns is multivariate normal with parameters that are stationary over time. Since the normal distribution is stable over addition, any portfolio of stocks will possess normally distributed returns. However, consistent findings (e.g., Fama, 1965, Teichmoeller, 1971, Praetz, 1972, Blattberg and Gonedes, 1974, Kon, 1984, and Ritchey, 1986) have indicated that returns are peaked and fat tailed relative to normal and most often positively skewed.
The observed returns seem to suggest a stochastic process composed of a mixture of distributions. There have been attempts in the literature to define this mixture. Mandelbrot and Taylor (1967) indicate a combination of normal stable distribution. Clark (1973) suggests a normal-lognormal mixture. Blattberg and Gonedes (1974) derive a student t from a normal gamma distribution. Merton (197 6) and Cox and Ross (197 6) introduce a return process that is a mixture of a continuous diffusion and a Poisson process. Kon (1984) proposes a discrete mixture of normal distributions to explain the significant kurtosis (fat tails) and the significantly positive skewness in the distribution of daily rates of return.
CONTENTS
ACKNOWLEDGEMENTS
LIST OF TABLES
LIST OF FIGURES
I. INTRODUCTION
II. LITERATURE REVIEW
- Efficient Market Hypothesis and Related Issues
- Does Information Matter to Investors
EMH When there are Hetrogenous Expectations
EMH When Information is Costly
Evidences Against a Fully Aggregating Market
The Information Structure Analysis
Market Microstructure Analysis
Security Analysts Related Studies
III. THEORETICAL MODEL
- Definition of the Time Subscript
Theories of Security Price Behavior
Assumptions of Investors' Behavior
Security Price Behavior When There is a Constant
Arrival of New Information and the Market is
Not Efficient
Testable Implications
Content of Information
Uncertainty of Information
IV. EMPIRICAL DESIGN
- Data
- Source and Derivation of Information Data
Source and Derivation of Stock Returns Data
Sample
Methodology
- Tests of the Relationship Between Stock
Returns and Content of New Information
Tests of the Relationship Between Volatility
of Stock Returns and Uncertainty of
Information
V. EMPIRICAL RESULTS
- Tests of the Relationship Between Stock Returns and Content of New Information
Method One: Using Observed Returns as the Dependent Variable
Method Two: Using Risk-Adjusted Returns as the Dependent Variable
Method Three: Portfolio Approach
Tests of the Relationship Between Volatility of Stock Returns and Uncertainty of Information
VI. SUMMARY AND CONCLUSION
- Summary of Results
Limitations of the Study
Suggestions for Future Research
ENDNOTES
REFERENCES
APPENDIX A THE DISTRIBUTIONS OF T-STATISTICS FOR THE ESTIMATED COEFFICIENTS OF EQ.4.9
APPENDIX B THE DISTRIBUTIONS OF T-STATISTICS FOR THE ESTIMATED COEFFICIENTS OF EQ.4.13
Download
PDF Ebook Information Structure And Stock Return Distribution; An Empirical Examination
Posted in :
