武汉市在校大学生数与经济增长的协整分析

近二十年来,我国高等教育规模不断扩大,高校在校生数不断增加,其中武汉市位居全国前列．为了研究高校在校生数对当地经济发展的影响,本文利用武汉市1992—2012年在校大学生人数数据,通过建立合适的模型,来定量探讨武汉市在校大学生数与经济增长的内在关系．结果表明,武汉市经济增长与武汉在校大学生人数模型之间存在着长期均衡关系且互为格兰杰因果关系．     

Measuring causality by taking the directional symbolic mutual information approach

We propose a novel measure to assess causality through the comparison of symbolic mutual information between the future of one random quantity and the past of the other.This provides a new perspective that is different from the conventional conceptions.Based on this point of view,a new causality index is derived that uses the definition of directional symbolic mutual information.This measure presents properties that are different from the time delayed mutual information since the symbolization captures the dynamic features of the analyzed time series.In addition to characterizing the direction and the amplitude of the information flow,it can also detect coupling delays.This method has the property of robustness,conceptual simplicity,and fast computational speed.     

An empirical study on information spillover effects between the Chinese copper futures market and spot market

This study employs a parametric approach based on TGARCH and GARCH models to estimate the VaR of the copper futures market and spot market in China. Considering the short selling mechanism in the futures market, the paper introduces two new notions: upside VaR and extreme upside risk spillover. And downside VaR and upside VaR are examined by using the above approach. Also, we use Kupiec’s [P.H. Kupiec, Techniques for verifying the accuracy of risk measurement models, Journal of Derivatives 3 (1995) 73-84] backtest to test the power of our approaches. In addition, we investigate information spillover effects between the futures market and the spot market by employing a linear Granger causality test, and Granger causality tests in mean, volatility and risk respectively. Moreover, we also investigate the relationship between the futures market and the spot market by using a test based on a kernel function. Empirical results indicate that there exist significant two-way spillovers between the futures market and the spot market, and the spillovers from the futures market to the spot market are much more striking.     

莆田市旅游业的发展与区域经济增长关系的实证研究

基于1996-2013年莆田市旅游业收入与地区GDP的相关数据,研究莆田市旅游业的发展与区域经济增长的关系.考虑到这些数据是时间序列数据,首先对这些序列进行平稳性分析,然后应用协整理论和Granger因果检验进行实证研究,并且建立了误差修正模型.结果表明,莆田市旅游业的发展与区域经济增长存在着长期稳定的关系和单向的Granger因果原因,即区域经济带动旅游业的发展影响较小,而旅游业的发展促进区域经济发展的影响较明显.     

基于计量经济模型的煤炭消费与经济增长关系实证研究——以世界主要煤炭消费国为例

经济的发展和能源(包括煤炭)的利用有着密不可分的关系.运用国际上研究能源消费与经济增长之间关系的重要工具-协整分析、格兰杰因果关系检验对世界主要煤炭消费国家(中国、美国、印度、俄罗斯、日本和南非)1981-2005的煤炭消费与GDP之间因果关系及其内在规律进行了分析和研究.实证研究结果表明,这些国家能源消费结构、经济政策不同,煤炭消费同经济增长之间的因果关系也不完全一致,中国、日本和南非存在经济增长对煤炭消费的因果关系,俄罗斯存在煤炭消费对经济增长的因果关系,而印度煤炭消费和经济增长之间则不存在因果关系.据此各国根据自身能源国情及国际比较形势,制定科学的能源战略与政策.     

房地产行业的数学建模分析

分别建立了需求模型,供给模型,房地产与其他产业的关系模型,房价模型及房地产发展趋势的预测模型,并对影响模型的相应因素做了简要的分析.     

中国房地产投资与国民经济关系结构变点研究

本文首先描述了房地产投资与GDP间关系在以2001、2008年为分界点的三个时期上的不同特征;其次用变结构协整模型侦测协整关系的结构变点;再次利用分段协整检验、邹氏断点检验和格兰杰因果检验对变点做进一步验证。结论如下:房地产投资与国民经济发展总体上存在长期协整关系,两者关系的发展呈现三段式:2001年之前两者尚未形成长期均衡关系,协整关系不显著,1998年到2001年是形成期;2001年之后两者形成了长期均衡关系,表现为协整关系比较显著;2008年之后协整关系脱离原均衡水平,向高位发展。加强对房地产行业的监督与调控,促使其与国民经济保持长期均衡,对经济社会持续、健康、协调发展意义重大。     

基于SD的城镇就业人数影响因子模型及其仿真

运用协整与Granger因果分析理论确定了国内生产总值、固定资产投资、国家财政收入、全社会消费品总额和第三产业产值为城镇就业人数的影响因子,运用系统动力学的理论与方法构建了城镇就业人数的系统动力学模型,在不同的投资和消费力度下对城镇就业人数进行了模拟。结果表明,消费相对投资而言更能提高就业水平,进而提出相应的对策及建议。     

Learning in Convolutional Neural Networks Accelerated by Transfer Entropy

Recently, there is a growing interest in applying Transfer Entropy (TE) in quantifying the effective connectivity between artificial neurons. In a feedforward network, the TE can be used to quantify the relationships between neuron output pairs located in different layers. Our focus is on how to include the TE in the learning mechanisms of a Convolutional Neural Network (CNN) architecture. We introduce a novel training mechanism for CNN architectures which integrates the TE feedback connections. Adding the TE feedback parameter accelerates the training process, as fewer epochs are needed. On the flip side, it adds computational overhead to each epoch. According to our experiments on CNN classifiers, to achieve a reasonable computational overhead–accuracy trade-off, it is efficient to consider only the inter-neural information transfer of the neuron pairs between the last two fully connected layers. The TE acts as a smoothing factor, generating stability and becoming active only periodically, not after processing each input sample. Therefore, we can consider the TE is in our model a slowly changing meta-parameter.