Extreme events have a material impact on return distributions and investment decisions. However, the role of event risks is understated in popular financial decision making approaches. This thesis includes event risks into investment decisions to improve global investment optimality. We examine event risks in two different but coherent financial settings: portfolio selection and corporate finance. In the portfolio selection setting, we focus on the incorporation of higher order information to capture the impact of event risks on portfolio construction. Higher order extensions are implemented on two main portfolio optimization methods: the classic framework of mean variance optimization and CAPM, and the stochastic dominance approach. We find that the inclusion of higher order information improves global portfolio optimality given the presence of event risks. As a special case, we combine the traditional applications of mean variance optimization and stochastic dominance analysis to examine the index efficiency of DJIA. We find that DJIA is efficient as a performance benchmark. In the corporate finance setting, we principally identified corporate name changes of M&As among the S&P 500 index, and examined how the name change events impact the return patterns for the acquirers and the targets. Conducting this corporate event study, we show that name changeevents substantially affect return dynamics, and that the abnormal return difference between name change events and non name change events is economically and statistically significant. Generally, our studies illustrate that the inclusion of event risks in decision processes brings important benefits to the asset allocation optimization.