标题： 期货市场限价订单簿交易量轮廓的重尾特征与实证分析 显示英文标题

标题： Heavy-Tailed Features and Empirical Analysis of the Limit Order Book Volume Profiles in Futures Markets

摘要： 本文针对限价订单簿随机结构的成交量分布属性提出了几个基本问题，并考虑了日内和隔日统计特征、不同交易所特性的影响以及市场参与者在不同资产领域中的影响。 本文旨在探讨以下问题： 1. 在限价订单簿的买入和卖出档位上是否存在统计证据表明重尾子指数分布的成交量分布出现在不同的层次上？如果是这样，这种现象是在日内还是隔日的时间尺度上发生？ 2. 在期货交易所中，重尾特征是否依赖于交易所（CBOT、CME、EUREX、SGX 和 COMEX）或资产类别（政府债券、股票和贵金属），并且它们是否出现在超高速（<1秒）或中等范围（1秒-10分钟）的高频数据中？ 3. 随机重尾成交量分布特征的存在是否以某种方式发展，从而能够反映或指示市场参与者的某些行为，例如高频算法交易、报价填充和日内价格发现？ 4. 是否存在统计证据表明需要在日内时间尺度上考虑限价订单簿成交量分布模型参数的动态行为？ 通过对前所未有的大量期货市场LO B数据集进行统计严格的验证，获得了每个问题方面进展的结果。 该数据集包括多个交易所、多种期货资产类别以及2010年的所有交易日，使用在买入和卖出档位上的5个深度（类型II）订单簿数据。 显示英文摘要

摘要： This paper poses a few fundamental questions regarding the attributes of the volume profile of a Limit Order Books stochastic structure by taking into consideration aspects of intraday and interday statistical features, the impact of different exchange features and the impact of market participants in different asset sectors. This paper aims to address the following questions: 1. Is there statistical evidence that heavy-tailed sub-exponential volume profiles occur at different levels of the Limit Order Book on the bid and ask and if so does this happen on intra or interday time scales ? 2.In futures exchanges, are heavy tail features exchange (CBOT, CME, EUREX, SGX and COMEX) or asset class (government bonds, equities and precious metals) dependent and do they happen on ultra-high (<1sec) or mid-range (1sec -10min) high frequency data? 3.Does the presence of stochastic heavy-tailed volume profile features evolve in a manner that would inform or be indicative of market participant behaviors, such as high frequency algorithmic trading, quote stuffing and price discovery intra-daily? 4. Is there statistical evidence for a need to consider dynamic behavior of the parameters of models for Limit Order Book volume profiles on an intra-daily time scale ? Progress on aspects of each question is obtained via statistically rigorous results to verify the empirical findings for an unprecedentedly large set of futures market LOB data. The data comprises several exchanges, several futures asset classes and all trading days of 2010, using market depth (Type II) order book data to 5 levels on the bid and ask.