Thesis (M.Sc.)--Chulalongkorn University, 2010

It has been empirically confirmed that media has a significant effect on the stock market. Mass media can alleviate information symmetry as well as informational frictions to a certain extent and affect security pricing even if it does not provide genuine news. The paper investigates this hypothesis by empirically studying the effect of media coverage on stock returns. Using non-regression and regression approaches, this paper also hopes to contribute to the expanding research within specific geographical region by investigating Thai data and examining the media effect in Thai stock market. The media effect found in the non-regression approach is a total opposite to the regression approach. However, the regression approach provides a stronger evidence to the media effect, as risk factors such as market, size and book-to-market factors are properly captured and controlled in CAPM and Fama-French three-factor model. It is found that using a media trading strategy that takes a long position on stocks with low media coverage and short position on stocks with high media coverage, stocks with high media coverage outperform stocks with low media coverage, even after controlling for firm characteristics such as size and book-to-market ratios. The result is a total opposite to the media effect found in the U.S.