Performance-weighted ensembles of random forests for predicting price impact

For any large player in financial markets, the impact of their trading activity represents a substantial proportion of transaction costs. This paper proposes a novel machine learning algorithm for predicting the price impact of order book events. Specifically, we introduce a prediction system based on ensembles of random forests (RFs). The system is trained and tested on depth-of-book data from the BATS and Chi-X exchanges and performance is benchmarked using ensembles of other popular regression algorithms including: linear regression, neural networks and support vector regression. The results show that recency-weighted ensembles of RFs produce over 15% greater prediction accuracy on out-of-sample data, for 5 out of 6 timeframes studied, compared with all benchmarks. Feature importance ranking is used to explore the significance of various market features on the price impact, finding them to be highly variable through time. Finally, a novel procedure for extracting the directional effects of features is proposed and used to explore the features most dominant in the price formation process.     

沪深300指数与股指期货价格变化关系的检验

在沪深300股指期货推出两年以后,为了验证期货市场的价格发现功能,本文利用VAR模型对沪深300股指期货和现货价格的每分钟收益率序列实际数据进行检验.实证检验结果显示期货价格变化对未来现货价格变化有显著影响,持续时间可达5分钟以上,而现货价格对期货价格变化没有显著影响.因此,沪深300指数期货市场已经发挥了较强的价格发现功能.同时,我们发现对股指期货和现货市场而言,VAR模型具有较强的稳定性,在利用历史数据预测未来指数价格变化中有较强的实用价值.     

The economics of online television: Industry development,aggregation, and “TV Everywhere”

Although still dominated by standard television, the online TV industry is growing rapidly. Entrants employ a range of business models, and we identify a prevalent tendency for leading providers to aggregate programming from a variety of different content owners. We focus on one form of content aggregation by multi-channel programming distributors (MPVDs) widely known as “TV Everywhere (TVE).” Following a brief taxonomy of TVE systems, we develop an economic model to show how this “free-with-authentication” (of MVPD subscribership) bundling practice can be explained as a price discrimination device intended to slow MVPD disconnections. We show that TVE bundling could also deter entry into the online TV market. We discuss the potential roles of horizontal and vertical integration of MVPDs and ISPs in online TV industry development, again focusing on TVE, and conclude with policy implications.     

The Well-Being Cost of Inflation Inequalities

In terms of well-being, how costly is inflation? To answer this question, empirical evaluations have typically studied average inflation rates at the national level, thus disregarding the role of inflation inequalities within a country. In this article, we relax the assumptions that heterogeneous consumers face homogeneous inflation rates, and study the correlation between price changes and self-reported satisfaction with living standards. We use newly available data from France and adopt two approaches. First, we focus on individually perceived inflation and use the internationally harmonized Opinion Price Index as a proxy for experienced inflation. Variations in perceived inflation help predict well-being differences among consumers, even when controlling for relevant sociodemographic factors, personality traits, and common method variance. We estimate their marginal impact to be higher than equivalent variations in nominal income. Second, we compare groups of consumers over time and find that changes in the price of a good disproportionately affect the relative well-being of those who consume it. The study shows that the well-being cost of the inflation crisis would be underestimated if looking at aggregate figures only.     

Antecedents of firm performance in emerging economies: Business groups,strategy, industry structure,and state support

A rich literature has investigated the antecedents of firm performance in developed economies, resulting in a consensus view that firm resources and strategy are the key determinants. Several arguments, however, suggest that in emerging economies other factors are more important for firm performance. This study analyzes the impact of firm strategy and industry structure as well as business group membership and state support on firm performance in an advanced emerging economy, Turkey. Using a data set compiled from a selection of the 1000 largest manufacturing firms in this country, the study employs several regression models to identify the main determinants of firm performance as measured by productivity and net profit margin. In contrast to studies of developed economies, the investigation finds that firm-related factors (competitive strategies) do not significantly influence performance; instead factors related to industry structure and business group membership are the strongest determinants of firm performance; further, state support interacts with business group membership and is positively related to productivity.     

浅谈企业社会责任——由“三鹿奶粉事件”想到的

"奶粉事件"又一次引起人们对企业社会责任的思考,企业要实现可持续发展的目标,保持长期的竞争优势,就必须承担相应的社会责任。     

出口退税的贸易效应研究 ——来自中国服装出口的经验评估

在我国历年出口退税政策调整中,服装产品退税率的调整频率和幅度在各行业中最为突出。以引力模型为基础,研究我国出口退税政策对服装出口的影响。实证结果表明,在控制贸易双方GDP、地理距离、货币相对购买力和国别特定因素条件下,出口退税政策调整对服装出口具有显著影响,提高出口退税率可以稳定或扩大出口,有效应对外需疲软。为消除出口骗税产生的虚假贸易数据的影响,以扣除贸易成本的进口国报告的进口额衡量我国出口贸易规模,实证结果再次印证,出口退税对服装出口有显著促进作用。     

战略视角决策框架研究:基于竞争战略和蓝海战略的整合分析

战略视角决策是战略制定和战略实施的隐含前提。战略视角包括以竞争战略为代表的结构主义战略视角和以蓝海战略为代表的重构主义战略视角。从挖掘理论本质出发,对比分析了两种战略视角的内涵和特点,采用理论演绎方法,揭示了两种战略视角整合情况下的产业内企业数量与行业平均利润间的长期关系、短期关系以及调整过程,基于演绎分析归纳了影响战略视角决策的3个关键因素,并提出了战略视角决策框架,为企业做出正确的战略视角决策提供了一定的理论支持。     

我国股指期货市场价格发现功能研究——基于上升、下跌和震荡状态下的分析

为了研究我国股指期货市场的价格发现功能,文章按照时变的思路,根据股指期货在不同市场环境下的作用和表现,将市场区别为上升、下跌和震荡三种情况来检验价格发现功能的差异。通过采用VEC模型、PT模型和IS模型进行对比分析,发现在大牛市和熊市时期,股指期货的所起到的作用会很明显,在价格发现中所占比例较大,为50-70%;而在股市平盘震荡时期,股指期货的价格发现能力要弱一些,只占20-30%比例。文中建议要加快发展我国股指期货市场,改善和优化目前的产品结构体系,减少对市场的不必要限制措施,使之成为更加规范和成熟的股指期货市场。     

囤积居奇的反垄断问题解析

当前,对企业囤积居奇行为的规制在我国遭遇了法律适应的问题,文章认为应依据《反垄断法》对企业囤积居奇行为进行规制。根据行动企业针对的目标市场不同可以将囤积居奇行为分成掠夺性囤积居奇和提高竞争对手成本性囤积居奇两种类型,对两者进行经济学分析可以得到两类囤积居奇行为对市场竞争及消费者福利所产生的影响存在一定差异。在对企业囤积居奇行为进行反垄断规制的实践中需遵循合理原则,依从四步骤对企业囤积居奇行为进行判定。