Essays in Financial Econometrics

The first chapter of this dissertation examines a contemporaneous relationship between transaction return and net order flow defined as a difference between buyer- and seller-initiated transactions using the method of identification through heteroskedasticity. The identification of the contemporaneous relation allows a decomposition of the return variance into two components driven by orthogonal return and flow shocks. The empirical results with the best bid and offer files of the E-mini S&P 500 futures contract illustrate intraday patterns of the contemporaneous relationship and the variances. Additionally, the results show that macroeconomic announcements increase the flow-related terms such as the variance driven by flow shock whereas they decrease liquidity risk measured by the fraction of the variance driven by flow shock to the total variance. Moreover, the results show that the interaction between trading activities, such as trading volume and the number of trades, and return variance differs between the two variance components but is the same when an announcement is released. These results suggest that it is necessary to consider both unique and common factors in variance components when modeling trading activity and return variance.;The second chapter, joint with Toshiaki Watanabe and Yasuhiro Omori, investigates the predictive performance of the realized stochastic volatility model of Takahashi, Omori, and Watanabe (2009), which incorporates the asymmetric stochastic volatility model with the realized volatility. Considering the well known characteristics of financial returns, namely heavy tails and skewness, the model is extended by employing a wider class distribution, the generalized hyperbolic skew Student's t-distribution, for financial returns. With the Bayesian estimation scheme via Markov chain Monte Carlo method, the model enables us to estimate the parameters in the return distribution and in the model jointly. It also makes it possible to forecast volatility and return quantiles by sampling from their posterior distributions jointly. The model is applied to quantile forecasts of financial returns such as value-at-risk and expected shortfall as well as volatility forecasts and those forecasts are evaluated by various tests and performance measures. Empirical results with the US and Japanese stock indices, Dow Jones Industrial Average and Nikkei 225, show that the extended model improves the volatility and quantile forecasts especially in some volatile periods.;The third chapter studies a new, simulation-based method to compute the news impact curve for stochastic volatility models. The simulation-based method, proposed by Takahashi, Omori, and Watanabe (2013), incorporates the joint movement of return and volatility, which has been ignored by the extant literature. News impact surface, a three dimensional plot of the tomorrow's volatility on today's news shock and today's volatility, illustrates the impacts of today's news shock and volatility shock separately. Simulation example shows a clear distinction between the news impacts computed by the simulation-based method and the conventional method. Moreover, the difference becomes larger as today's volatility level moves away from its unconditional mean. These results support the importance of considering the today's news shock and today's volatility shock separately when estimating the news impact curve for stochastic volatility models.