基于改进Transformer模型的文本摘要生成方法

传统的文本摘要方法，如基于循环神经网络和Encoder-Decoder框架构建的摘要生成模型等，在生成文本摘要时存在并行能力不足或长期依赖的性能缺陷，以及文本摘要生成的准确率和流畅度的问题。对此，提出了一种动态词嵌入摘要生成方法。该方法基于改进的Transformer模型，在文本预处理阶段引入先验知识，将ELMo(Embeddings from Language Models)动态词向量作为训练文本的词表征，结合此词对应当句的文本句向量拼接生成输入文本矩阵，将文本矩阵输入到Encoder生成固定长度的文本向量表达，然后通过Decoder将此向量表达解码生成目标文本摘要。实验采用Rouge值作为摘要的评测指标，与其他方法进行的对比实验结果表明，所提方法所生成的文本摘要的准确率和流畅度更高。     

Sparse estimation of dynamic principal components for forecasting high-dimensional time series

We present the sparse estimation of one-sided dynamic principal components (ODPCs) to forecast high-dimensional time series. The forecast can be made directly with the ODPCs or by using them as estimates of the factors in a generalized dynamic factor model. It is shown that a large reduction in the number of parameters estimated for the ODPCs can be achieved without affecting their forecasting performance.     

基于底线预警的国土空间生态修复规划实施监督体系研究

国土空间生态修复是推进生态文明建设的重大举措，构建基于底线预警的国土空间生态修复规划实施监督体系是完善国土空间规划“一张图”信息系统的重要组成部分和政策支撑。通过对生态修复规划实施监督体系建设的研究，把握底线控制与监测预警机制，对接国土空间生态修复业务管理各项需求，搭建“五梯度四体系”的生态修复规划动态监测实施监督信息系统，构建以“七个一”为主要内容的生态修复规划底线预警监测评估机制，并与各层级国土空间规划“一张图”实施监督信息系统进行衔接，实现对生态修复规划实施的动态监测和预警。     

The income elasticity gap and its implications for economic growth and tourism development: the Balearic vs the Canary Islands

ABSTRACT

The Balearic and the Canary Islands are two well-known tourism-led economies. They both experienced a tourism boom during the same decades, and, hence, they developed a similar productive-mix. Nevertheless, there are strong economic differences between the two regions. While the Balearic Islands enjoy a high GDP per capita, the Canary Islands show a more modest performance. The results of a panel data regression confirm our hypothesis that they differ substantially in terms of income elasticity of tourism. It is two times higher in the Balearic Islands than in the Canaries, which indicates the first is perceived as a more luxurious destination. Furthermore, the results of a dynamic computable general equilibrium model show that the Canaries would converge in GDP per capita with the Balearic Islands if they attracted tourists with a similar profile as the latter.     

Measuring industrial upgrading: applying factor analysis in a global value chain framework

ABSTRACT

A key question for promoting international competition is how to improve the position of countries and industries in global value chains (GVCs). The first step is to properly measure industrial upgrading in GVCs. This is not a trivial issue because upgrading has not been defined unambiguously. Several authors have used different (and sometimes related) measures, all of which indicate certain aspects of upgrading. Rather than trying to find the single, ultimate measure of upgrading, we propose a different approach. We examine the multidimensionality of industrial upgrading, using eight indicators in factor analysis. Four of the eight indicators adopt the GVC perspective and include, for example, the growth of the share in value-added exports. We provide three quantitative dimensions of industrial upgrading: process upgrading, product upgrading, and skill upgrading. With these dimensions, we compare and analyze the upgrading of different countries and industries using the World Input–Output Database.     

The impact of foreign direct investment on productivity: New evidence for developing countries

We provide new empirical evidence on the relationship between inward foreign direct investment (FDI) and total factor productivity (TFP) growth using cross-country data for 51 developing countries over the period 1984–2010. Our results suggest a weak direct effect of FDI on TFP growth but, after accounting for the roles of human capital and institutions as contingencies in the FDI-TFP growth relationship, we find a robust FDI-induced productivity growth response dependent on these ‘absorptive capacities’. However, the relevance of the human capital contingency effect diminishes when the effect of institutions is also considered, which suggests that improving institutions is relatively more important than human capital development for developing countries to realise productivity gains from FDI.     

Minorías étnicas foráneas y determinación salarial en China

Si bien algunos estudios observan que las minorías étnicas urbanas en China perciben salarios más bajos que la mayoría han, otros no constatan la existencia de esa brecha. Para resolver esta contradicción, los autores proponen desglosar las experiencias laborales por grupo étnico minoritario. El análisis de las minorías étnicas sobre un amplio conjunto de datos indica que las «minorías foráneas», como los tibetanos y los grupos túrquicos, sufren una penalización salarial significativa cuando se controlan las covariables, pero no ocurre lo mismo con la totalidad de las minorías. Estas conclusiones son robustas en varias especificaciones y tienen importantes implicaciones teóricas y en materia de políticas.     

山西省乡村休闲旅游地空间分布特征及影响因素研究

［目的］通过对山西省乡村休闲旅游地的空间分布特征及影响因素进行研究，旨在为山西省的乡村旅游发展提供合理的参考建议。［方法］选取山西省213处国家级和省级乡村旅游示范点为研究对象，运用最临近点指数对其空间分布类型进行判别，并运用地理集中指数和地理联系率对其分布均衡程度和地理联系程度进行考量，通过核密度图进一步分析其空间分布指向，在此基础上对影响因素进行分析。［结果］山西省乡村休闲旅游地空间分布呈集聚分布的特点，且与经济规模和人口密度在空间上的均衡、配合程度较高。［结论］（1）山西省乡村休闲旅游地包括以太原和晋中为主的中部密集区和以临汾、运城、晋城为主的南部密集区，且都以各地级市为中心向周边扩展，大多数示范点分布于城郊地区； （2）经济发展、人口密度、资源禀赋、区位因素和社会因素是影响山西省乡村休闲旅游地分布的主要因素，其中农业科技类、农事体验类和特色村镇类多分布于经济和人口密度较高的地区。     

区域制造业空间效应及生产性服务业发展模式选择

通过空间计量模型考察生产性服务业发展模式与制造业集聚的空间溢出效应后发现:生产性服务业专业化水平及多样化水平与制造业集聚呈倒U型关系,同时城市化水平会对服务业发展模式与制造业集聚关系产生影响。分区域模型研究发现,生产性服务业对制造业全要素生产率影响因区域经济发展水平、产业层次、要素禀赋等因素有所差异,东部地区制造业全要素生产率与生产性服务业呈线性关系,适合选择多样化发展模式;中部地区呈倒U型关系,现阶段中部地区生产性服务业更适合选择多样化发展模式;西部地区生产性服务业发展水平与制造业全要素生产率呈正U型关系,生产性服务业专业化和多样化发展都将促进制造业全要素生产率提升。