Decision trees unearth return sign predictability in the S&P 500

Technical trading rules and linear regression models are often used by practitioners to find trends in asset returns. However, these models typically neglect interaction terms between the lagged daily directional movements. We propose a decision tree forecasting model that has the flexibility to capture arbitrary interaction patterns. To study the importance of interaction terms, we construct a binary Markov process with a deterministic component that cannot be predicted without interaction terms between the lagged directional movements. We show that some tree based strategies achieve trading performance significant at the 99% confidence level on the S&P 500 over the past 20 years, after adjusting for multiple testing. The best strategy breaks even with the buy-and-hold strategy at 21 bps in transaction costs per round trip. A four-factor regression analysis shows significant intercept, and correlation with the market. The directional predictability is strongest during the bursts of the dotcom bubble, financial crisis, and European debt crisis. The return sign predictability during these periods confirms the necessity of interaction terms to model daily returns.     

福建省生态公益林碳汇计量

按照分层抽样方法在福建省抽取6个样本县,在样本县按照立地条件和主要生态公益林树种分别抽取成熟林样地,应用蓄积量换算因子法,计算出6个样本县的碳汇量和不同树种样地的碳汇量。计算结果表明:生态公益林各主要树种单位面积年均碳汇量从高到低依次为桉树、木麻黄、杉木、马尾松和阔叶树。     

Analysis of a network structure of the foreign currency exchange market

We analyse structure of the world foreign currency exchange (FX) market viewed as a network of interacting currencies. We analyse daily time series of FX data for a set of 63 currencies, including gold, silver and platinum. We group together all the exchange rates with a common base currency and study each group separately. By applying the methods of filtered correlation matrix we identify clusters of closely related currencies. The clusters are formed typically according to the economical and geographical factors. We also study topology of weighted minimal spanning trees for different network representations (i.e., for different base currencies) and find that in a majority of representations the network has a hierarchical scale-free structure. In addition, we analyse the temporal evolution of the network and detect that its structure is not stable over time. A medium-term trend can be identified which affects the USD node by decreasing its centrality. Our analysis shows also an increasing role of euro in the world’s currency market.     

上市公司股利分配倾向研究:基于数据挖掘方法的分析

股利分配是上市公司的一项十分重要的财务决策。而影响股利分配的因素有许多,国内现有文献一般是采用多元统计回归建模方法对其进行实证分析。本文应用数据挖掘中的决策树方法寻找股利分配的原因,这是文章的创新点。从决策树来看,现金股利分配不仅受股本结构、公司规模、当年可分配利润、成长机会及经营风险等客观财务因素的影响,而且与控股股东性质存在显著的影响,且国家控股股东分配倾向低于民营控股股东。     

两阶段实物期权的供应链模型研究

在一个单期、两级供应链中供应商和销售商共同开发投资一种新型产品,在此情况下,本文分别对不考虑期权和引入期权两种情况进行分析:考虑期权的情况下,在投资和销售两阶段分别引入投资延迟期权和销售量担保期权,通过对模型分析与求解发现,销售商通过购买期权将风险部分转移给供应商,供应商获得风险补偿的同时,由于承担了更多的风险将谨慎投资,从而使供应链整体风险得到有效控制,并且实现资金的有效配置。     

基于集对故障树方法的电网安全运行风险评估模型

本文首先对电网安全运行风险进行分析,并结合实际情况,运用故障树法构建电网安全运行模型;然后,利用集对分析的概念及其运算法则推导出集对与门算子和集对或门算子;接着,采取集对分析方法,得出具有确定性与不确定性的各个底事件的联系度;最后,以算例表明本文所提出的集对故障树分析法是可行的,既能实现对电网安全运行风险的度量,又避免了以往采用概率度量的不准确性,能更好地描述电网运行风险的真实程度。     

长沙市民文化旅游决策规则分析

随着文化旅游愈益重要,人们的文化旅游决策的影响因素是一个值得特别重视的问题。通过对长沙市居民进行问卷调查,获得1068份有效问卷,形成数据库,采用决策树技术分析,发现文化旅游者具有三级决策规则,分别对应＂文化底蕴＂、＂地方风俗＂和＂月收入＂。使用决策树建立决策规则模型,从年龄和职业角度对三级规则的人口构成进行了分析。     

运用高阶累积量和SVM的调制自动识别

针对数字信号调制模式识别问题,提出了运用高阶累积量和二叉树支持向量机（SVM）进行 自动识别的算法。该算法首先使用信号的四阶、六阶、八阶累积量构造了5个新的分类特征 ,然后利用二叉树支持向量机分类器实现了8种信号的有效分类。仿真结果表明,该算法优 于直接多类分类支持向量机算法,在信噪比大于5 dB时,识别率达到90%以上。     

基于改进C5.0的元胞自动机模型研究——以土地利用问题为例

元胞自动机模型是通过局部的相互作用及变化来模拟全局的变化,是复杂结构来源于简单系统的重要体现。元胞自动机模型的核心是转换规则的获取,获取转换规则的方法有多种。建立一种基于改进C5.0决策树的元胞自动机模型,基于属性值的相似性对C5.0决策树算法进行改进,以期提高元胞自动机的模拟精度,该模型适用于多种复杂系统的模拟问题研究。最后以土地利用问题为例演示了模型的应用。     

Development of automatic fault tree synthesis system using decision matrix

Fault trees synthesis, the basis for fault tree analysis (FTA), serves as a powerful tool for risk analysis. It has become a trend to accomplish computer-assisted fault tree synthesis in the field of system safety engineering because conventional manual construction of fault trees can be extremely time-consuming and vulnerable to human errors. This paper expounds upon a fault tree synthesis information system (FTSIS) developed by means of decision matrix for the purpose of its application to process plants, where the objective system is decomposed into a series of system components whose cause-effect models are constructed and stored in the relational database and transferred into decision matrix. The fault tree is synthesized automatically after the decision matrix is fully searched in FTSIS, the availability of which has been verified after it was put into effect successfully.