The main purpose of this thesis is to test whether the capital asset pricing model (CAPM) can be applied to the Korean stock market. To avoid some statistical problems the grouping procedure was used. Also the beta stability was tested over tweleve and thirty monthly observations. Since beta estimates over thirty months were more stable than those over twelve months, betas were estimated over thirty months for the grouping procedure.
The time series regressions of the portfolio excess returns on the market portfolio excess returns indicated that high-beta securities had significantly negative interceps and low-beta securities had significantly positive intercepts, contrary to the predictions of the traditional form of the CAPM originally developed by sharpe and Lintner. Also the cross-sectional regressions of the mean excess returns on the portfolios against the estimated betas indecated that the intercept and slope were not consistent with the traditional form of the CAPM.
Then the assumption that riskless borrowing and lending opportunities are available was relaxed and the resulting two factor model was tested by deriving a minimum-variance, unbiased linear estimator of the returns on the beta factor. The cross-sectional regressions of the mean total returns on the portfolios against the estimated betas indicated that the intercept and slope were very close to the predictions of the two factor model. Thus it is concluded that the two factor model can be applied to the Korean stock market.
