INTERACTION TERMS IN POISSON AND LOG LINEAR REGRESSION MODELS

This paper develops a difference‐in‐semielasticities (DIS) interpretation for the coefficients of dichotomous variable interaction terms in nonlinear models with exponential conditional mean functions, including but not limited to Poisson, Negative Binomial, and log linear models. We show why these interaction term coefficients cannot be interpreted as a DIS or semielasticity in the same manner as continuous coefficients, which has been overlooked by some empirical researchers. Then we show how interaction terms can be easily transformed into a DIS and derive the asymptotic distribution of this estimator. We illustrate the discrepancy between the interaction term coefficient and the DIS using an empirical example evaluating the relationship between employment, private health insurance and physician office visits. Our results can be applied in treatment effect models when the outcome variable is logged and the dichotomous variables indicating treatment participation and the post‐treatment time period.     

Measuring the spatial price differences in China with regional price parity methods

The measurement of prices has been an important field in economics, and a spatial price index is very useful in comparing the standards of living and well-being across regions in a country. This paper intends to measure the regional price parities (RPPs) across different provincial areas in China with an urban sample of 140 goods and services in 2015 according to the framework of the International Comparison Program (ICP) methodology. The results show that the RPPs that were estimated with commonly used approaches, such as the Gini–Éltetö–Köves–Szulc (GEKS), Geary–Khamis (GK) and weighted country product dummy (WCPD), are only slightly different. The RPPs of three regions including Guangdong, Shanghai and Zhejiang are greater than 1 (with Beijing = 1), while the other 27 regions are all lower than 1, which represents price levels that are less than Beijing. In the extrapolation of the RPPs from 2000 to 2014, a significant descending trend is found for the standard deviation series of the RPPs over time that shows that the price differences across regions decreased during the extrapolating period. This finding provides evidence of a phenomenon of spatial price converging in China. Finally, a study on the deflation of provincial aggregates with the RPPs reveals that the spatial price adjustment will change the ranks or relative importance of different regions in the country. Especially, the measurement of income inequality proves that the Gini coefficients of provincial income deflated by the RPP^GEKS are all lower than the Gini coefficients of unadjusted incomes.     

Fast Kernel Smoothing of Point Patterns on a Large Network using Two‐dimensional Convolution

We propose a computationally efficient and statistically principled method for kernel smoothing of point pattern data on a linear network. The point locations, and the network itself, are convolved with a two‐dimensional kernel and then combined into an intensity function on the network. This can be computed rapidly using the fast Fourier transform, even on large networks and for large bandwidths, and is robust against errors in network geometry. The estimator is consistent, and its statistical efficiency is only slightly suboptimal. We discuss bias, variance, asymptotics, bandwidth selection, variance estimation, relative risk estimation and adaptive smoothing. The methods are used to analyse spatially varying frequency of traffic accidents in Western Australia and the relative risk of different types of traffic accidents in Medellín, Colombia.     

Has the crisis affected Spanish investment strategy abroad? A spatial panel data approach

This paper investigates whether the Spanish investment strategy abroad has been affected by the Great Recession. Applying a panel spatial Durbin model for two sample periods, pre‐crisis (1996–2007) and crisis (2008–14), our findings indicate that Spanish foreign direct investment (FDI) strategy has been indeed markedly altered by the global economic crisis. Complex‐vertical FDI motives prevailed over the pre‐crisis period, whereas horizontal FDI did so over the crisis. These results are robust to the use of sectoral FDI data and alternative specifications of the spatial weight matrix.     

斜拉-刚构组合体系桥零号块空间应力分析

以河谷汉江公路大桥斜拉桥主桥部分作为依托工程,针对其零号块空间受力问题,采用有限元软件midas FEA建立零号块有限元模型并进行相应分析.以施工过程中成桥运营阶段最大弯矩工况、最大剪力工况以及施工阶段最大双悬臂工况为控制计算工况,分析3种工况下零号块最不利内力情况.研究表明,在现有的设计条件下,结构受力可满足规范要求,相应的设计细节可为类似的工程提供借鉴.     

Spatial econometric analysis of spatial general equilibrium

A structural spatial econometric model for nine regions of Israel is estimated using non-stationary spatial panel data during 1987–2015. The model focuses on the relation between regional markets in labour, housing and capital when there is imperfect internal migration between regions, when capital is imperfectly mobile between regions, and when building contractors operate across regions. Since the regional panel data are non-stationary, the econometric methodology is based on spatial panel cointegration. The estimated model is used to simulate the temporal and spatial propagation of regional shocks induced, for example, by regional policy (land for housing, regional investment grants). Impulse responses are temporally and spatially state dependent. They are also highly persistent because of longevity in housing and capital.     

The Effect of Investment,LFA and Agri-environmental Subsidies on the Components of Total Factor Productivity: The Case of Slovenian Farms

The effect of subsidies on the performance of farms has received a great deal of attention in the literature, although results are inconclusive. Furthermore, much of the related literature examines the effect of subsidies only on technical efficiency (TE). We examine the effect of different types of subsidies on the different components of total factor productivity (TFP) in Slovenian agriculture over the period 2006–2013. We first estimate a Random Parameter Stochastic production frontier model. Then, based on the estimates of this model, we calculate and decompose the TFP index into TE, scale efficiency and technological change. Third, we apply combined difference-in-difference and a matching estimator to examine the effect of investment, less favoured area (LFA) and agri-environmental (AE) subsidies on the different components of TFP. In our case, these subsidies are found to have no significant effect on either TFP or on its components.     

The Long‐Term Consequences of Cultural Distance on Migration: Historical Evidence from China

This paper uses a unique dataset of long‐run migration in China to provide new measures of cultural distance based on biological evidence. We utilise this information to examine the effect of cultural distance on migration in China from the Song Dynasty until the twentieth century. Our findings show that culture has strong effects on migration between regions when controlling for other socio‐economic and geographical factors. Population flow is greater where there are fewer cultural differences. This finding not only provides quantitative empirical evidence on linking cultural distance and migration but also leads to a better understanding of the mechanisms resulting in migration for much of China's history.     

金融集聚、购房压力对科技创新的时空影响机制:以粤港澳大湾区为例

综合考虑时间与空间两个维度,识别出影响粤港澳大湾区科技创新的关键因素,旨在促进大湾区创新资源整合和创新水平提高。基于粤港澳大湾区11个城市2006-2016年面板数据,采用面板门限模型和空间计量模型来分析区域内金融集聚、购房压力对科技创新的非线性影响机制及空间溢出效应,在此基础上比较行业异质性影响。结果显示:粤港澳大湾区金融集聚对科技创新存在显著的正向空间溢出效应,购房压力对科技创新存在一定程度的负向空间溢出效应。从整个金融业来看,大湾区内金融集聚对科技创新具有显著的促进作用,而购房压力对科技创新没有显著的统计意义上的关系。从金融子行业来看,保险业集聚对科技创新具有显著的正向促进作用,而银行业集聚和证券业集聚对科技创新均没有明显影响。金融(各子行业)集聚与购房压力的互动均对科技创新产生显著的抑制作用;金融集聚、购房压力对科技创新的统计意义上影响具有门限效应,呈现出非线性影响机制。在低、中度购房压力区域内,金融集聚对科技创新具有显著负向影响;在高购房压力区域内,金融集聚对科技创新具有显著正向影响;与此同时,在低、高度金融集聚区域内,购房压力对科技创新没有明显的统计意义上的影响;但在中度金融集聚区域内,购房压力对科技创新具有一定促进作用。地区研发投入、经济发展水平、教育水平、政府扶持政策对科技创新产生积极促进作用,但存在结构性差异。     

城市轨道交通对周边住宅价格的空间效应研究

以合肥轨道交通1号线为例，使用传统特征价格模型和空间计量模型对比分析轨道交通对周边住宅价格的影响。研究表明合肥市1号线对周边住宅的价格具有显著的增值作用，且住宅价格的空间效应显著。空间滞后模型显示住宅与轨道交通站间距离每增加1 km，住宅价格下降6.09%，低于传统模型的8.6%。这说明住宅价格的空间效应会使学者们高估城市轨道交通对周边住宅价格的增值作用，即加入空间效应影响后的研究结果更加真实、可靠。