Traditional portfolio theory suggests that, in equilibrium, only the market risk is priced in the cross-section of expected stock returns. However, if the market is not perfect and investors are constantly changing investing behaviors based on their perceptions about future market outlook, then non-traditional risk factors could potentially provide significant power of describing the expected stock returns. This dissertation has two essays on the pricing of volatility, in which the market is not assumed to be frictionless or perfect. Essay 1 focuses on the pricing of individual volatility in penny stocks. Empirical results show that individual volatility plays an important role in describing the average cross-sectional returns of penny stocks. Resorting to the rolling portfolio approach, evidences indicate that portfolios consisting of penny stocks with high individual volatilities, on average, earned much higher returns than portfolios consisting of penny stocks with low individual volatilities. This effect is statistically significant when multiple factors are controlled simultaneously. Essay 2 focuses on the pricing of the market volatility among individual stocks. Following the rolling portfolio method, Essay 2 constructs portfolios that consist of individual stocks with various market volatility exposures. Traditional risk factors such as market beta, size, book-to-market, and momentum are controlled …

Scholars and investors have studied the value premium for several decades. However, the debate over whether risk factors or biased market participants cause the value premium has never been settled. The risk explanation argues that value firms are fundamentally riskier than growth firms. At the same time, the behavioral explanation argues that biased market participants systematically misprice value and growth stocks. In this paper, I use the implied cost of equity capital to capture all risks that investors demand a premium and sort stocks into risk quantiles. The implied cost of equity capital is estimated using models proposed by Gebhardt et al., Claus and Thomas, Ohlson and Juettner-Nauroth, and Easton. I find that value stocks have higher implied cost of equity capital and lower forecasted earnings growth while growth stocks have lower implied cost of equity capital and higher forecasted earnings growth. More importantly, even within the same risk quantile, the value premium still exists. The results suggest that risk and behavioral factors simultaneously cause the value premium. Furthermore, by decomposing the holding period return, I find that adjustments in valuation ratios caused by negative earnings surprises for growth firms and positive earnings surprises for value firms at least partially …