本项目利用金融高频数据中所包含的市场波动率的精确信息，提出一个新的离散时间下的期权定价模型Realized Heston-Nandi，并系统的研究其理论性质和应用。该模型是在Realized GARCH以及经典的Heston-Nandii期权定价模型的基础上扩展而成的，并且兼具这两类模型的优点。新模型和Heston-Nandi 模型一样，能够推导出欧式期权定价公式的解析形式, 而且还保留了Realized GARCH模型简单优美的数学结构和易于估计的特性。同时，由于已实现波动率测度包含波动率的精确信息，在快速变化的市场环境中仍能较好的预测未来波动率，因此能更好的拟合实际期权数据。本项目还将研究该模型的几种重要扩展形式及其在金融衍生品市场中的应用，比如波动率长记忆性在期权定价中的影响，Realized Heston-Nandi模型在定中国市场期权定价的实证表现。

By using the accurate information of high frequency data about underlying volatility, this project proposes a new discrete-time option pricing model Realized Heston-Nandi and systematically studies its theory and applications. The new model is based on the combination of the Realized GARCH and the classical Heston-Nandi model and shares the advantages of the two models. Like Heston-Nandi model, our new model generates the closed-form pricing formula for European options. It also keeps the nice mathematical structure and the estimation simplicity of Realized GARCH. Meanwhile, due to the accurate information contained in realized volatility, our new model is able to predict future volatility accurately in a fast-changing market situation and therefore provides a better fit for option prices. The project also studies several important extensions of the model and their applications in financial derivatives markets, such as, the impact of long memory of volatility on option pricing; the empirical performance of Realized Heston-Nandi model in Chinese options market.