在对高维度、多属性、非同步限价指令簿流数据的统计特征与影响因素研究基础上，提出快速极大似然聚类方法，检测指令簿流数据实时状态特征。对Hawkes过程进行扩展，使其自激发和互激发计数过程与指令簿状态过程完全耦合、相互影响，构建自适应状态相依Hawkes过程，模拟限价指令簿事件的动态演化。采用多层次排队理论对限价指令簿中交易执行过程进行研究，推导出给定状态下最优控制轨迹，在此基础上给出限价指令簿市场微观结构变量对价格冲击和交易成本的影响函数。针对传统Q学习算法必须逐点估计最优策略，难以估计处于两个行动点之间的行动，以及需要存储空间随着状态空间的维度呈指数增长的不足，设计出双重深度Q学习的方法求解交易者的最优交易策略。对本项目提出的双重深度Q学习最优交易执行方法的有效性和经济学解释进行分析，对限价指令簿中的信息、学习和交易行为研究，研究限价指令簿中的信息、主体学习对市场效率和流动性等影响。

Optimal trade execution is an important problem faced by essentially all traders. Based on the features in complex adaptive system of financial market, and high-dimensional, multi-feature and non-synchronous stream data in limit order book，We propose the application of a high-speed maximum likelihood clustering algorithm to detect temporal financial market states, using correlation matrices estimated from intraday market microstructure features, then we study statistical aspects of state-dependent Hawkes processes as a candidate governing process for limit order book events.At the same time, in order to better understanding the dynamic evolution of limit order book, We will model an electronic limit order book as a multi-class queuing system and formulate and solve a problem of limit and market order placement to optimally buy a block of shares over a short, predetermined time horizon. Using the structure of the optimal execution policy, we identify microstructure variables that affect trading costs over short time horizons and propose a resulting microstructure-based model of market impact costs，Meanwhile, we develop a variation of Deep Q-Learning to estimate the optimal actions of a trader. The model is a fully connected Neural Network trained using Experience Replay and Double DQN with input features given by the current state of the limit order book, other trading signals, and available execution actions, while the output is the Q-value function estimating the future rewards under an arbitrary action.By introducing double deep Q-Learning into an order-driven limit order market, we will study the effectiveness and rationality of the agent learning by examining the price discovery, the impact of learning on market liquidity, including liquidity supply and consumption, order profit, bid-ask spread, and information efficiency.