含空间自回归误差项的空间动态面板模型的检验与模拟

研究目标：探究含空间自回归误差项的空间动态面板数据（SDPD）模型选择和假设检验结果的水平扭曲、检验功效。研究方法：构建含空间自回归误差项的SDPD模型的空间Hausman、LM和LR检验统计量,选择蒙特卡洛模拟进行分析。研究发现：含空间自回归误差项的SDPD模型的各类检验统计量的大样本性质良好;时空滞后项对空间Hausman检验结果的影响比空间滞后项显著;条件LMλ|α、LRλ|α检验是含空间自回归误差项随机效应SDPD模型的最优检验统计量,时空滞后项对LM、LR检验结果的影响比自回归误差项更显著。研究创新：解析了SDPD模型中各空间关联项系数的波动对模型检验精度的影响。研究价值：探究含空间自回归误差项SDPD模型的选择问题,为实证提供理论依据。     

京津冀城市群区域经济效率的测度及影响研究

运用2003~2015年京津冀面板数据,采用空间面板数据计量模型,研究京津冀经济增长的空间溢出效应,测算要素投入及TFP对经济增长的贡献。结果表明:京津冀各区市县之间的经济增长及要素之间存在较强的空间依赖性和空间溢出效应。研究进一步发现资本投入依然是京津冀经济增长的主要源泉,京津冀经济增长依然是要素投入的结果;资本投入对经济增长的贡献较大,劳动力对经济增长的贡献相对较小,大部分区市县全要素生产率对经济增长的贡献为负。研究的政策涵义在于:促进京津冀区域经济一体化发展,需要合理引导要素在不同区域间流动,提升要素在不同区域间有效配置;其次,需要进一步提高技术进步水平,发挥人力资本的优势,充分发挥劳动力的效率。     

改革开放以来中国经济增长动力转换的时空特征

基于中国1978~2012年的省级面板数据,采用检验后的超越对数生产函数的随机前沿模型,利用两套资本存量核算方法,对中国经济增长动力的来源及其时空特征进行分析,并对2008年金融危机的4万亿元投资政策进行了效率评价。研究结果表明,中国属于典型的投资主导型经济,资本投入是中国经济增长持续稳定的最主要来源,TFP贡献率呈现逐年下降的趋势;中国经济增长动力由改革开放初期的资本、劳动力和TFP三驾马车平衡拉动,形成了现阶段的资本投入与TFP反向角力态势;区域经济差距主要源于资本投入与TFP双重差异,但TFP差异是最重要因素;4万亿元经济刺激政策下中国经济复苏属于典型的“投资主导型复苏”,是以牺牲中国生产率为代价的,TFP在2008年后呈现断崖式下降,平均拉低中国TFP达0.23~0.32个百分点。     

中国装备制造业全要素生产率增长的波动与异质性

研究目标：中国装备制造业及其细分行业TFP增长率的测定与分解。研究方法：构建超越对数形式的随机前沿生产函数模型,测算中国装备制造业及其细分行业TFP增长率,并对其进行分解。研究发现：中国装备制造业TFP增长率不但在区域和省份之间存在差异,而且存在明显的内部行业异质性;技术进步和配置效率变化分别是提高和阻碍装备制造业TFP增长率的主要原因和障碍;除技术进步均为正值外,技术效率变化、规模效率变化和配置效率变化在装备制造业各行业中异质性相当明显;装备制造业及其细分行业仍未从真正意义上实现由粗放型向集约型增长方式的转变。研究创新：中国装备制造业细分行业TFP增长率及其异质性。研究价值：为装备制造业转型与升级提供经验证据。     

考虑时间特征的客观AHP判断矩阵构造方法

研究目标:为将面板数据中指标的时间维度纳入客观AHP判断矩阵的构造中,因此对时间相依性客观AHP赋权方法展开研究。研究方法:从时间序列自相关性、指标数值转移余量和随机过程转移概率三个维度构造判断矩阵,对比四种赋权方法的优缺点,通过数值仿真评估四种方法的可行性与实用性,以2012～2020年中国31个省份的数字普惠金融指数再合成为例比较各方法测算结果的异同。研究发现:不同赋权类型下各时期内子系统是否通过一致性检验存在区别,不同赋权类型下样本综合指数的数值大小、涨幅趋势和波动情况既有区别又有相同之处,不同赋权类型下样本综合指数的时空特征不尽相同。研究创新:基于时间维度给出了客观AHP判断矩阵的构造方法。研究价值:在AHP方法中加入了时间维度,使客观AHP评价方法更适用于面板数据。     

非平衡面板数据的全要素生产率测算方法

研究目标:在广义DEA的理论框架下对传统DEA-Malmquist指数方法进行修正，从而使非平衡面板数据的全要素生产率测算结果更为合理。研究方法:应用广义DEA方法的基本原理，将传统DEA-Malmquist指数方法的理论基础从“自评体系”拓展到“他评体系”,给出一种修正的DEA-Malmquist指数方法，并将这些方法应用于2005～2018年中国西部地区经济全要素生产率的测算中。研究发现:基于广义DEA理论的修正DEA-Malmquist指数方法有效消除了数据缺失对Malmquist指数测算的影响，在非平衡面板数据的全要素生产率测算方面更具合理性和稳定性。研究创新:在广义DEA的理论框架下实现了非平衡面板数据全要素生产率的有效测算。研究价值:为全要素生产率的测算提供了全新思想，同时也增强了DEA方法在处理非平衡面板数据方面的能力。     

中国区域生产率的变动及分解——基于Malmquist生产力指数的实证研究

文章基于省际面板数据，运用DEA下的MalmquiSt指数方法测算并分解了社会主义市场经济体系建立以来的全要素生产率（TFP）变化，并对变化原因进行了解释。     

区域创新活动的空间效应及影响因素研究

研究目标：探究区域创新活动的空间效应及影响因素。研究方法：采用1998～2014年中国30个省份的面板数据,运用空间计量经济学,建立静态与动态空间面板模型,并从地理特征与社会经济特征两方面构建5个空间权重矩阵进行实证分析。研究发现：区域创新活动具有正向空间相关性,并呈现明显的空间集聚现象;地理特征与社会经济特征对区域创新活动均产生影响,其效果以社会经济特征更大;利用动态空间面板模型进行估计优于静态空间面板模型。研究创新：根据地理距离空间矩阵进行的空间相关性检验与R&D资本存量新估算方法使模型结果更准确;交通水平距离空间矩阵构建合理。研究价值：进行区域创新空间相关性分析时,可准确选择适当模型、估算方法及影响因素。     

中国区域资本配置效率及区域资本形成影响因素的实证分析

本文构建了32组包含中国各行业近11万个数据的大型面板数据集,分行业系统测算了我国各区域的资本配置效率,并对区域和省际资本形成影响因素进行了系统的比较分析。研究发现:从短期来看,我国各地区的资本形成更加依赖于金融机构贷款的支持;绝大多数省份的资本形成存在自我约束、收敛的机制;不同省份间金融要素对固定资本形成的影响存在较大差异,说明我国地区之间存在一定程度的资本流动障碍。     

基于随机前沿面神经网络的TFP贡献率测度

针对全要素生产率(TFP)贡献率测度的问题,研究了用神经网络来测量TFP贡献率的方法和用随机前沿面来测量TFP贡献率的方法;指出了这两种方法在测算中存在的缺陷。并在此基础之上,提出了运用随机前沿面模型构造出随机前沿面神经网络来测量TFP贡献率,从而实现了利用随机前沿面神经网络模型进行TFP测度贡献率。全部计算过程都通过了编程计算,收进我们自己的软件系统DASC。     

异质性能源消费与经济增长的非线性动态驱动机制

研究目标：测度改革开放近40年来,中国异质性能源消费与经济增长的非线性动态驱动机制。研究方法：基于煤炭、石油、天然气、电力消费以及GDP年度数据,运用马尔科夫区制转移因果(MSC) 模型开展实证研究。研究发现：经济增长能够促进煤炭、石油、天然气和电力等异质性能源消费的提高,其中,石油和天然气消费的增加能够推动经济增长。经济增长对煤炭和石油消费的驱动作用持续期较长,对天然气和电力消费的驱动作用持续期较短,四种能源消费对经济增长驱动作用的时间长度大致相同。近年来,石油消费能够表现出对经济增长的非线性动态驱动作用,经济增长能够对石油和煤炭消费发挥非线性动态驱动作用。在金融危机时期,难以表现出经济增长对能源消费的单向时变因果影响,其中,煤炭、石油和天然气消费无法表现出对经济增长的单向时变因果影响,而电力消费对经济增长存在单向时变因果影响。研究创新：基于MSC模型,判断不同时段内异质性能源消费与经济增长的时变因果关系,进而揭示两者之间的非线性动态驱动机制。研究价值：为中国完成能源产业结构转型以及构建新时代能源产业体系提供经验证据。     

Energy consumption and economic growth: evidence from non-OPEC oil producing states

According to the US EIA (2009, www.eia.doe.gov), out of the 15 largest oil producing nations in the world, 7 are not OPEC members, namely Brazil, Canada, China, Mexico, Norway, the Russian Federation, and the United States of America (USA). This paper investigates the causal relationship between energy consumption and economic growth for these non-OPEC oil producing countries. Real GDP per capita is used to measure economic growth; whereas; energy consumption is represented by four sub-variables (electric power, oil, natural gas, and coal energy). Using a panel data covering (1969–2009), this study employs the Pedroni (Econometric Theory, 20, 597–627, 2004) approach to determine cointegration and the (Econometrica, 55, 251–276, 1987) two-step procedure to explore short and long run causal effects. The results suggest that there are long run relationships between the real GDP, labour force, real capital, oil consumption, electricity consumption, gas consumption and coal consumption. Further analyses show that real GDP and oil consumption Granger cause real gross capital formation in the short run; real gross fixed capital and electricity consumption cause oil consumption in the short run; and also oil consumption and gas consumption cause electricity consumption in the short run.     

Renewable and non-renewable energy consumption,environmental degradation and economic growth in Tunisia

The aim of this document is to investigate the dynamic relationship between economic growth, renewable energy consumption, energy consumption and CO₂ emissions in Tunisia over the period 1990–2015. Unit root tests and co-integration test was used in order to detect the order of stationary and to test the existence long run links between the used variables. We apply the Granger causality test and VECM model to discover the short and long run links between the variables. Results have shown a bidirectional causal relationship between energy use and CO₂ emissions. Economic growth affects CO₂ emission in the short and long run. While there is a unidirectional links running from energy use to economic growth at short run. The paper shares best practices from Tunisia in terms of efficient use of renewable energy policy enablers, which may be contextualized in other emerging economies in order to keep sustainability and to achieve the green economy.     

辽宁电力能源消费与经济增长的实证分析

随着社会经济快速发展和人口迅猛增长,各国的经济增长均面临不同程度的能源瓶颈约束,文章基于Granger因果检验、协整分析等计量模型和方法,分析了辽宁省电力能源消费与经济增长之间的内在联系,通过分析得出如下结论:辽宁省电力消费与经济增长之间存在长期协整关系,电力能源消费引起生产总值GDP的变化,是经济增长的单向Granger因果关系。最后针对辽宁省就如何加快电力能源结构调整,推动能源产业转型升级,促进电力与经济协调发展方面提出了具体对策建议。     

山东省经济增长与煤炭能源消费关系的实证研究

本文基于1990-2012的时序数列,利用协整分析、误差修正模型及格兰杰因果检验,实证检验了山东省经济增长与煤炭能源消耗之间的长期均衡和短期波动情况,并利用脉冲响应分析了两者的动态响应路径。     

Electricity consumption and economic growth empirical evidence from Pakistan

The present article uses the Autoregressive Distributed Lag (ARDL) bounds testing procedure to identify the long run equilibrium relationship between electricity consumption and economic growth. Toda Yamamoto and Wald-test causality tests have identified the direction of the causal relationship between these two variables in the case of Pakistan in the period between 1971 and 2008. Ng-Perron and Clement-Montanes-Reyes unit root tests are used to handle the problem of integrating orders for variables. The results suggest that the two variables are in a long run equilibrium relationship and economic growth leads to electricity consumption and not vice versa.     

Individual new energy consumption and economic growth in China

We document the one-way relationship between individual new energy consumption and economic growth in China through the autoregressive distributed lag (ARDL) model from 2004 to 2017. Our results show that individual new energy consumption has a positive effect on economic growth. Moreover, the urbanization rate, import and export trade volume and foreign direct investment all affect the individual new energy consumption in the short run. The outcome of the causality test reveals a one-way Granger causal relationship from individual new energy consumption to economic growth, from the urbanization rate, and from the import and export trade volume to new energy consumption.     

Energy consumption, CO2 emissions, and economic growth: evidence from Indonesia

Energy policy-makers in Indonesia are interested in the causal relationship between energy consumption, CO₂ emissions, and economic growth. Therefore, this paper attempts to analyze the short- and long-run causality issues between energy consumption, CO₂ emissions, and economic growth in Indonesia using time-series techniques. To this end, annual data covering the period 1965–2006 are employed and tests for unit roots, co-integration, and Granger-causality based on an error-correction model are applied. The results show that there is a bi-directional causality between energy consumption and CO₂ emissions. This means that an increase in energy consumption directly affects CO₂ emissions and that CO₂ emissions also stimulate further energy consumption. In addition, the results support the occurrence of uni-directional causality running from economic growth to energy consumption and to CO₂ emissions without any feedback effects. Thus, energy conservation and/or CO₂ emissions reduction policies can be initiated without the consequent destructive economic side effects.     

The path to renewable energy consumption in the European Union through drivers and barriers: A panel vector autoregressive approach

Renewable energy consumption brings sustainable economic growth and pollution reduction. Despite the worldwide increase in renewable energy consumption, global energy-related carbon dioxide emissions are rising and there are still considerable differences in the share of renewable energy consumption in national energy portfolios. These concerns require further effort at the policy level, especially by countries that make extensive use of energy imports. These countries could improve their lack of energy independence by using renewable energy sources and leveraging a few factors to facilitate their transition. This study aims to investigate renewable energy consumption drivers, focusing on the role of socio-technical (rather than economic) aspects such as policy stringency, lobbying, public awareness, and education. We employ a panel vector autoregressive model in first differences to test the complex dynamic relationships among renewable energy consumption, policy stringency, lobbying, public awareness, and education, controlling for variables such as per capita income and import levels, for 12 European Union net energy importing countries. Results show that the positive income effect prevails in the influence of the level of carbon dioxide emissions (negative) on renewable energy consumption, despite the latter being more significant in countries with higher levels of education. Increasing energy needs push traditional sources towards complementarity with renewable energy consumption, implying a positive lobbying effect. Public awareness is not enough to facilitate the transition to renewable energy consumption. By contrast, policy stringency has positive direct and indirect effects on renewable energy consumption, suggesting that the approach adopted by the European Commission in the recent Green Deal is a step in the right direction. Moreover, as shown, policymakers are able, through renewable energy consumption, to generate a decrease in carbon dioxide emissions and electricity production from oil, gas, coal, and nuclear sources in the first instance, but also in net energy imports, even if at a later stage.     

Oil price shocks,geopolitical risks,and green bond market dynamics

This research explores the causal relation among oil price, geopolitical risks, and green bond index in the United States from December 2013 to January 2019. Unlike the conventional linear model specification used in earlier works, we evaluate causal relations based on Granger-causality in quantile analysis. Our empirical results reveal unidirectional Granger-causality from geopolitical risk to oil price at the extreme quantiles. We also observe a significant bi-directional causality from oil price to green bond index for the lower quantiles. Findings also reveal causality from geopolitical risk to green bond index in the lower quantiles of the distribution. Therefore, knowledge of these causal relationships can help policy makers to evaluate and implement effective policies to prevent sudden and substantial oil price shocks and geopolitical risk.