Date of Award

2011

Document Type

Thesis

Publisher

Edith Cowan University

Degree Name

Bachelor of Business Honours

School

School of Business

Faculty

Faculty of Business and Law

First Supervisor

David Allen

Abstract

Many studies have tested the CAPM and the Fama and French model in the Australian security market using the Ordinary Least Squares (OLS) method. However, this regression method just focuses on the relationship between means in the dataset, and equity market usually has some extreme situations in the tails. In this study, quantile regression will be used as well as OLS to provide a more comprehensive picture. This research will also compare the domestic and overseas indices in testing the CAPM and the Fama and French model. A twenty-year data sample composed of the 50 largest companies' equity returns will be analyzed in the first-pass regression. In the second-pass regression, results estimated in the first-pass regression under OLS and quantile regression will be used as the independent variables, to check the relationship between different factors and the subsequent equity return. Seventeen portfolios will be sorted according to the methodology of Fama and MacBeth (1974) to generate a cross-sectional dataset. The regression power and testing results will be compared between different regression methods and datasets. The empirical results generally demonstrate that the international dataset has performed better than the domestic dataset; and factors estimated by quantile regression in the first-pass regression have better explanatory power for the subsequent equity returns in the second-pass regression.

Included in

Econometrics Commons

Share

 
COinS