The expansion of electronic communications networks over the past decade has had a tremendous impact on the efficiency of stock market trading. The Nasdaq exchange has even added additional trading sessions that extend beyond the open market period. While the volume during these pre-market and after hours periods is dwarfed by the open market session, there is still considerable trading for highly active stocks. This dissertation analyzes the relationship between extended hours and open market volatility as well as addresses the impact of extended hours volatility on traditional option pricing models.

In Chapter 1, I demonstrate the effects of incorporating extended hours realized volatility into autoregressive models of daytime volatility. Using the AR(1) model as a benchmark, I show that the pre-market and after hours regressors contribute significant in-sample explanatory power above and beyond lagged daytime volatility. Moreover, they also produce statistically significant improvements in one day out-of-sample forecasts by as much as 28%. These efficiency gains are robust to an optimally chosen AR(4) model as well as the heterogeneous autoregressive model.

In Chapter 2, I propose a stochastic volatility option pricing model wherein the parameters change deterministically on a daily basis in accordance with the trading environment. I calculate option prices by deriving the characteristic function and estimate the parameters jointly under the objective and risk-neutral probability distributions. Results show that the deterministic regime change model improves the simpler classic version on which it is based by about 39%. This is largely due to the regime changing model's better ability to price short-dated out-of-the money options as well as long-dated in-the-money options.

In Chapter 3, we apply modern asset pricing techniques to value the Social Security Administration's open group liability. We treat the growth rates of net aggregate tax receipts and benefits as rates of return on derivative securities. We price these securities by spanning the growth rates on traded financial assets. We find that depending on the set of traded securities, the open group liability may be as much as four to five times larger than the official estimate of $15.1 trillion.