Empirical and theoretical applications of real options models in finance

This thesis involves theoretical and empirical work addressing two themes: (1) volatility dynamics of early-stage firms with jump risk and stage-clearing; (2) the adoption of management accounting innovation.;The first chapter uses a hand-collected data set of early-stage firms in the U.S. biotech industry to study the volatility dynamics of early-stage firms, which usually have a major large Research and Development (R&D) project that requires multi-stage investment. I document two effects: first, the success (failure) of R&D efforts within each stage decreases (increases) the uncertainty or volatility level of future returns: jump-risk effect; second, as firms survive each investment stage and are becoming mature, the uncertainty level of their future returns should eventually decrease in later investment stages that lead to maturity: stage-clearing effect..;In the second chapter, to explain jump-risk effect and the stage-clearing effect, I model an early-stage firm as a sequence of nested call options with jumps that lead to a mature firm. The jump-risk effect arises because an early-stage firm is modeled as a compound call option with jumps on the underlying cash flows; the volatility of an early-stage firm at each stage is determined by the option elasticity to the underlying cash flows. If a downside (upside) jump happens, the value of the underlying cash flows decreases (increases), which makes the option elasticity go up (down). As a result, the compound call option becomes more risky (less risky). The stage-clearing effect arises because as the firm exercises its option to continue investment, the new option that the firm enters into will eventually become a less risky option.;The third chapter employs a real option approach (ROA) to study the decision of Activity-Based Costing (ABC) adoption under uncertainty. We find that the optimal entry threshold for adoption obtained by the ROA is higher than that obtained by the net present value (NPV) method. In contrast, the optimal exit threshold for discontinuation obtained by the ROA is less than that obtained by the NPV method. The difference between these two methods is primarily caused by the option value of waiting before implementing the entry (exit) project in the ROA.