Characteristic approach to high risk low return puzzle in SET

Abstract

Theoretically investing in high risk asset generates high return. However, actual payoff in market is in the opposite side. In several empirical studies documented high risk stocks paid low return even lower than the portfolio of the lowest volatile stocks. I examine whether the cause of this anomaly is the lottery stocks. Lottery stocks which are low price, high skewness and high kurtosis might be desired by many investors even institutional investors. The assumption is that this high demand of lottery stocks might push the price up and cause low average return finally. Therefore, I set up the hypothesis these risky stocks are lottery stocks. I collect monthly total return, market value, book to market and price of SET listed and delisted stocks in 2002 to 2019 to construct asset pricing model by mimicking portfolio as Carhart 4 factors. I calculate excess return, market risk premium, smb, hml, umd and lottery factors similar to Carhart 4 factors but used skewness, kurtosis and price to form mimicking portfolio factors instead. I classify SET stocks into 5 portfolios sorting by volatility focusing on the highest volatile whether lottery factors affect their return. Nevertheless, there is no evidence enhancing the hypothesis that they are lottery stocks. However, although not all lottery factors can explain the return of high risk stocks, but skewness factor can. I can imply that high risk stocks in Thai stocks market is partially lottery stocks. In additional, I found this asset pricing model is fitter to explain return and generate less alpha after GRS testing which can be implied that this lottery factor model is informative or meaningful to the return.