Spatial Weight Research Articles

The impact of logistic environment and spatial spillover on agricultural economic growth: An empirical study based on east, central and west China.

To vigorously promote the integrated development and mutual adaptation of agriculture and logistics is an important way to realize agricultural modernization and rural revitalization. Along with the policy support of agricultural industry chain and the steady rise of the demand for agricultural products market, the total amount of agricultural product logistics continues to increase, and the growth rate remains stable. With the booming development of "Internet + agriculture" and e-commerce platform, agricultural logistic market welcomes a new round of development opportunities, reaching several trillion yuan. Compared with the developed countries, our agricultural product logistics is still far behind. At present, only 15% of vegetables and fruits and 30% of fresh meat have professional logistic transportation, while most of the rest are still in the state of local and primitive. The gap of logistic environment construction and logistic elements determines the difference in final benefit of agricultural products. The purpose of this study is to find out the influence of logistic elements on regional agricultural economic growth, and whether the influence between neighboring regions presents "the same prosperity", "the same loss" or "sharing weal and woe". Based on the panel data of the statistical yearbook of 31 provinces in China from 2005 to 2020, the spatial Durbin model was constructed under the spatial weight matrix of economic distance and economic geographical distance to conduct empirical analysis, and the internal factors of logistics industry, factor spillover effect and its impact on agricultural economic growth were studied. Results showed that: (1) considering economic distance factor, the spatial coefficient of the time-fixed effects model passed the significance test in eastern China. Considering economic geographic distance factors, the individual and double fixed effect models passed the significance test in central China, and all models passed the significance test in western China. (2) From the perspective of logistic infrastructure, AVLFA, HM, TN and RM had a positive effect on the growth of agricultural economy in eastern China, but LIAV is on the contrary. AVLFA had a positive effect on agricultural economic growth, but TN was on the contrary in central China. In western China, LIAV and TN promoted agricultural economy while HM and RM held back it. From the perspective of the volume of logistics activity, both eastern and central regions did not pass the significance test, but FA was tested by the double fixed effect model and showed negative in western China. From the perspective of control variables, FU, AO and PT all promoted agricultural economy in eastern China, and FU and AO did the same in central and western China. PT was invalid in central regions and hindered agricultural economy in western China, which was different. From the perspective of spatial spillover effect decomposition, the eastern region presents "one prosperity and all prosperity, and sharing weal and woe", while the central and western regions present "one prosperity and all prosperity, and one lost and all lost". At last suggestions as formulating the overall plan for the development of regional logistics, paying attention to regional differences and promoting coordinated development of logistics and agriculture in light of local conditions, and paying attention to the spatial spillover effect of elements were put forward.

Graph convolutional networks with learnable spatial weightings for traffic forecasting applications

How to select a suitable spatial weighting scheme for convolutional graph neural networks (ConvGNNs) is challenging. In this study, we propose a ConvGNN, termed learnable graph convolutional (LGC) network, which learns spatial weightings between a road and its k-hop neighbours as learnable parameters in the spatial convolutional operator. A dynamic LGC (DLGC) network is further proposed to learn the dynamics of spatial weightings by explicitly considering the temporal correlations of spatial weightings at different times of the day. A multi-temporal DLGC (MTDLGC) network is developed for forecasting traffic variables in road networks. Results of case study suggest that the MT-DLGC network can achieve higher prediction accuracy than other state-of-the-art baselines. Both LGC and DLGC networks can be used as general spatial weighting schemes for baselines with better forecasting performance than existing spatial weighting schemes, e.g., graph attention. The source code of this study is available publicly at https://github.com/Mayaohong/MTDLGC.

Exploration of spatiotemporal heterogeneity and socio-demographic determinants on COVID-19 incidence rates in Sarawak, Malaysia

BackgroundIn Sarawak, 252 300 coronavirus disease 2019 (COVID-19) cases have been recorded with 1 619 fatalities in 2021, compared to only 1 117 cases in 2020. Since Sarawak is geographically separated from Peninsular Malaysia and half of its population resides in rural districts where medical resources are limited, the analysis of spatiotemporal heterogeneity of disease incidence rates and their relationship with socio-demographic factors are crucial in understanding the spread of the disease in Sarawak.MethodsThe spatial dependence of district-wise incidence rates is investigated using spatial autocorrelation analysis with two orders of contiguity weights for various pandemic waves. Nine determinants are chosen from 14 covariates of socio-demographic factors via elastic net regression and recursive partitioning. The relationships between incidence rates and socio-demographic factors are examined using ordinary least squares, spatial lag and spatial error models, and geographically weighted regression.ResultsIn the first 8 months of 2021, COVID-19 severely affected Sarawak’s central region, which was followed by the southern region in the next 2 months. In the third wave, based on second-order spatial weights, the incidence rate in a district is most strongly influenced by its neighboring districts’ rate, although the variance of incidence rates is best explained by local regression coefficient estimates of socio-demographic factors in the first wave. It is discovered that the percentage of households with garbage collection facilities, population density and the proportion of male in the population are positively associated with the increase in COVID-19 incidence rates.ConclusionThis research provides useful insights for the State Government and public health authorities to critically incorporate socio-demographic characteristics of local communities into evidence-based decision-making for altering disease monitoring and response plans. Policymakers can make well-informed judgments and implement targeted interventions by having an in-depth understanding of the spatial patterns and relationships between COVID-19 incidence rates and socio-demographic characteristics. This will effectively help in mitigating the spread of the disease.

Spatial nonlinear effects of urban vitality under the constraints of development intensity and functional diversity

Urban vitality is the comprehensive characterization of the activities and processes in the city, which is influenced not only by environmental factors but also by the vitality of surrounding areas. Thus, the spatial effect of urban vitality can be understood under the constraint of the built environment. Taking Xiamen Island as a case, multisource big data are used to measure urban vitality, development intensity, and functional diversity. Spatial weight matrices with different distances are constructed. Through the spatial autoregressive threshold model, the threshold values of development intensity and functional diversity are estimated, and the spatial nonlinear effects of urban vitality based on different spatial weight matrices are analyzed. Results show that: (1) There is a significant spatial spillover effect of urban vitality. The spatial spillover effect is influenced by the built environment conditions and shows the characteristics of nonlinear growth with the change in development intensity and functional diversity; (2) the spatial spillover effect of urban vitality is significantly enhanced with the expansion of influence distance; (3) The development intensity has double thresholds. With the improvement of development intensity, the spatial spillover effect of urban vitality becomes stronger and more significant. The functional diversity has a significant single threshold value. The discussion of functional diversity should be carried out on the premise of density. The conclusion provides a theoretical basis for the optimization of urban vitality and built environment.

YOLOv5-ASFF: A Multistage Strawberry Detection Algorithm Based on Improved YOLOv5

The smart farm is currently a hot topic in the agricultural industry. Due to the complex field environment, the intelligent monitoring model applicable to this environment requires high hardware performance, and there are difficulties in realizing real-time detection of ripe strawberries on a small automatic picking robot, etc. This research proposes a real-time multistage strawberry detection algorithm YOLOv5-ASFF based on improved YOLOv5. Through the introduction of the ASFF (adaptive spatial feature fusion) module into YOLOv5, the network can adaptively learn the fused spatial weights of strawberry feature maps at each scale as a way to fully obtain the image feature information of strawberries. To verify the superiority and availability of YOLOv5-ASFF, a strawberry dataset containing a variety of complex scenarios, including leaf shading, overlapping fruit, and dense fruit, was constructed in this experiment. The method achieved 91.86% and 88.03% for mAP and F1, respectively, and 98.77% for AP of mature-stage strawberries, showing strong robustness and generalization ability, better than SSD, YOLOv3, YOLOv4, and YOLOv5s. The YOLOv5-ASFF algorithm can overcome the influence of complex field environments and improve the detection of strawberries under dense distribution and shading conditions, and the method can provide technical support for monitoring yield estimation and harvest planning in intelligent strawberry field management.

Analysis of Pre-post Covid-19 Influence in Bangka Belitung Islands Province: Socio-Economic Aspects in 7 Regencies/Cities

This study will analyze the variables of economic growth, poverty rate, unemployment rate, and the Human Development Index using situational trend analysis of the 4 factors before - after Covid-19 and spatial autocorrelation to see the inter-regional linkages as well as the distribution pattern of the observed data. The purpose of the study was: 1) to find out the description of the situation of the 4 factors before – after Covid-19; and 2) knowing the spatial autocorrelation based on Moran’s index with the spatial weighting matrix (WIJ). Based on the results of the analysis, the rate of economic growth in each district/city of Prov. Bangka Belitung tends to have a downward trend in the period 2018–2020 except Kab. Bangka and Pangkalpinang City. The variable rate of economic growth also did not have a spatial autocorrelation during the 2018-2021 period, which was indicated by the distribution pattern of the 2018-2019 data spreading (the Moran index was negative) and 2020-2021 was clustered (the Moran index was positive). Unemployment variable in all districts/cities of Prov. Bangka Belitung had a significant upward trend in the 2018-2021 period. The variable percentage of unemployment also did not have a spatial autocorrelation, which was indicated by a data distribution pattern that spreads over the 2018-2021 time period. The poverty variable in all districts/cities of Prov. Bangka Belitung had a downward trend in the 2018-2021 period. The variable percentage of poverty also does not have a spatial autocorrelation, which was indicated by a data distribution pattern that spreads over the 2018-2021 observation period (the Moran index was negative). Variable Human Development Index in all regencies/cities prov. Bangka Belitung has an upward trend during the 2018-2021 timeframe. The HDI variable also does not have a spatial autocorrelation, but the data distribution pattern tends to collect during the 2018-2021 time period (positive Moran index).
 Keywords: economic growth, poverty, unemployment, human development, spatial autocorrelation, Covid 19

Use of remote sensing-based pressure-state-response framework for the spatial ecosystem health assessment in Langfang, China.

Land use/land cover changes are occurring at an unprecedented rate and scale because of the economic development that has jeopardized the ecosystem's health. Ecosystem health should be studied and monitored at spatiotemporal scale to promote sustainable development and ecological civilization. The goal of this study was to assess the spatial ecosystem health of Langfang at the city and administrative levels using city's regional characteristics. Remote sensing-based pressure-state-response (PSR) framework, analytical hierarchy process (AHP), and principal component analysis (PCA) were utilized for spatial ecosystem health index (SEHI) formulation, indicator weighting, and indicator selection in several epochs (1990, 2003, 2013, and 2021), respectively. SEHI was formulated by combining subindices of pressure, state and response. The spatial ecosystem pressure index (SEIP) identified that the pressure was increasing on the ecosystem. In contrast, the spatial ecosystem state index (SEIS) pointed out an improvement in the state of the ecosystem since 1990. The worst state of the ecosystem was observed for the year 2013. The spatial ecosystem response index (SEIR) indicated that the response of the ecosystem towards the exerted pressures and states remained variable; however, it was reasonably good in 1990. All the administrative units of Langfang were associated with a healthy score for the spatial ecosystem health index (SEHI) for 1990 (pre-industrialization epoch), while the SEHI significantly reduced in 2013 (industrialization epoch) however improved for the later epochs (circular economy and ecological civilization epoch). The SEHI was moderately healthy for Dachang, Dacheng, Guan, Guangyang, and Yongqing while relatively healthy for the remaining administrative units in 2021. SEHI identified that spatial health has been improving since 2003 though not reaching the 1990's level for Langfang. Therefore, efforts should be focused on minimizing pressure and stabilizing the state to improve the spatial ecosystem health of Langfang. The developed SEHI can assist policymakers in analyzing regional health, identifying development strategies, driving environmental restoration, and quantifying needed changes.

Is artificial intelligence associated with carbon emissions reduction? Case of China

With the rapid development of modern information technology, the widespread application of artificial intelligence (AI) technology has had an increasingly critical impact on China's industrial development in recent years. However, few research focuses on the impact of AI development on China's carbon emissions (CEs). Using a data set of 30 Chinese provinces during the period 2006–2019, this study investigates the impact of AI development on CEs and explore potential influencing mechanisms via spatial Durbin model (SDM). It shows that AI development effectively contributes to CEs reduction and the reduction effect remains consistent over varied spatial weights. In terms of the mechanism analysis, technique and structure effects of AI reduce CEs. The heterogeneity analysis reveals that AI reduces CEs through spatial spillover effect with the effect being stronger in central and western China. Finally, based on the findings of this study, some recommendations are provided to promote the development of AI in China and the reduction of CEs.

The Spatial Autoregressive Panel Data Model with Spatial Moving Average Errors

This paper advocates the wider use of the spatial autoregressive (AR) panel data model with spatial moving average (MA) errors, individual and time effects, and different spatial weight matrices for each spatial lag. We demonstrate the practical relevance of this model, derive and investigate the asymptotic properties of a simple quasi maximum likelihood within estimator when is large and is finite, and provide an empirical example explaining military expenditures in 144 countries over the period 1993–2007.

Sensitivity analysis of Bayesian nonparametric spatial crash frequency models

This study focused on the sensitivity analysis of Bayesian nonparametric spatial models which combine the strengths of spatially structured random effects and the Dirichlet mixture to account for the unobserved heterogeneity of crash counts. Various evaluation criteria were employed to compare the performance of models with varying spatial weight matrices and precision parameters (alpha). The results demonstrate that there exists strong correlation among the posterior number of clusters, alpha, the fraction of variation explained by the spatial random effect, and different evaluation criteria. Even though the increased upper bound value of alpha does not necessarily lead to the enhanced model performance, the models with the full flexibility to choose the desirable amount of clustering tend to perform better than those with limited flexibility due to smaller allowable mass components. Compared with the precision parameter, no obvious trend is illustrated for the different evaluation criteria along the varying spatial weight matrices.

SPATIAL DURBIN MODEL OF UNEMPLOYMENT RATE IN CENTRAL JAVA

Unemployment is a labor problem that is often faced by developing countries like Indonesia. The number of unemployed in Indonesia has fluctuated from year to year, including in Central Java Province. One of the efforts made to overcome this problem is to know the factors that influence unemployment. The region effect greatly affects the open unemployment rate. Modeling involving area effects is very precise, one of which is the Spatial Durbin Model (SDM). In this study, modeling of the open unemployment rate was carried out using a spatial approach in each district/city in Central Java. The models used in this study are Ordinary Last Square (OLS), Spatial Auto Regressive (SAR), Spatial Error Models (SEM), Spatial Durbin Model (SDM), Spatial Error Durbin Model (SDEM). The five methods were evaluated using the Akaike Information Criteria (AIC). The spatial weighting used in this study is Queen Contiguity. Based on the smallest AIC value (115.42), the best method in this study is HR. Meanwhile, the significant factors are the percentage of labor force participation rate (X1), the number of poor people (X4), the lag of economic growth, the lag of poverty, and the lag of the district/city minimum wage

Water‐surface infrared small object detection based on spatial feature weighting and class balancing method

AbstractInfrared imaging is widely used due to its penetration capability to operate under many weather or lighting condition. However, due to the far distance of aerial view, feature blur, and the scarcity of aerial infrared data, the detection of small infrared targets on the water surface remains a challenging problem. In response to the problem of unclear features, we propose the spatial feature weighting method based on 2D Gaussian distribution. This method increases the weight of the target area by adaptively adjusting the feature activation. Secondly, for the problem of rare aerial perspective infrared data, we propose the cross‐spectral data migration method. By introducing the domain difference loss function to optimize the pseudo‐label selection process, the range of target domain distribution is expanded, and the adaptability of the detector is improved. Finally, in response to the problem of underfitting caused by category imbalance in transfer learning, we propose the class balancing method that effectively reduces the false detection. Extensive experiments were conducted on both benchmark datasets and the self‐built dataset to evaluate the effectiveness and robustness of our method. The proposed method was evaluated with different models and various scenarios, and the results demonstrated the effectiveness.

Synergetic Development Measure of Airport Groups Composite System and Its Influencing Factors Analysis: Some Evidence from China

Whether airport groups develop synergistically has become a concern for the aviation industry. Existing studies have neglected the dynamic comprehensive evaluation and analysis of influencing factors of synergy degree. Taking the synergistic development of airport groups as the research perspective, we integrate the dynamic comprehensive evaluation model into the traditional synergy model. From two dimensions of airport development and regional economy, we measured the static synergy degree of the Chengdu-Chongqing airport group and the dynamic synergetic development degree on the overall time series. Further, using a spatial econometric model, we construct a distance economy-nested spatial weight matrix so as to explore the influence of each indicator on the synergy degree of the composite system. The results show that the overall synergy of the Chengdu-Chongqing airport group is above 0.5 in terms of static synergy degree. The clustering spectrum shows that the Chengdu-Chongqing airport group can be divided into three categories and basically forms the characteristic of “two main and multiple auxiliary airports.” In terms of dynamic synergetic development degree and comprehensive evaluation value, the change speed of synergy degree is relatively smooth in the early stage and fluctuates in the later stage, and the overall dynamic evaluation value of the Chongqing regional airports is high. The effects of the explanatory variables of the airport group composite system on the synergy degree show more direct effects than spillover effects, and the direct effects are all positive feedback effects, while the negative spillover effects of the explanatory variables cannot be ignored.

Green finance, the low-carbon energy transition, and environmental pollution: evidence from China.

With 2008-2020 China's provincial panel data as the research object, we use the entropy weight method to compute green finance and pollution, and we empirically study the scope of the spillover effects and threshold impacts of green finance on environmental pollution using the spatial Durbin model and threshold regression model. The findings are as follows: first, under different spatial weight matrices, green finance has a spatial spillover effect on pollution. And the effect boundary is approximately 500km. Second, with the low-carbon energy transition, technological progress, and green finance as threshold variables, there are a single threshold, single threshold, and double threshold, respectively, that have significant threshold effects. Third, the spillover and threshold effects of green finance on pollution differ by region. The indirect effect is negative in the eastern region and positive in the central and western regions. There is a single threshold effect of low-carbon energy transition and green finance in the western region, a single threshold effect of green finance in the central region, and a single threshold effect of technological progress in the eastern region. On this basis, we put forward. specific policy recommendations.

Multi-scale telecoupling effects of land use change on ecosystem services in urban agglomerations --A case study in the middle reaches of Yangtze River urban agglomerations

A systematic understanding of the influences of land use change/land cover change (LUCC) on ecosystem services (ES) is significant for maintaining sustainable development of ecosystems. Previous studies on the effects of LUCC on ES have been discussed at multiple dimensions and scales, but there is still a lack of systematic studies on the influences of LUCC and its telecoupling effects on ES at different spatial scales and in different levels of urban agglomerations from a systemic perspective. Therefore, this paper endeavours to fill in the blanks. On the basis of quantifying the spatial and temporal evolution characteristics of LUCC and ES in the middle reaches of Yangtze River urban agglomerations (MRYRUA) and its three major metropolitan areas (C-Z-T metropolitan; Wu-Han Metropolian; Poyang Lake Metropolian), a spatial self-lagging of X model (SLX) based on the gravitational spatial weight matrix is constructed to explore this problem. The results show that the overall land use intensity of the MRYRUA has increased by 0.07% from 1990 to 2020, with the largest increase in the C-Z-T metropolitan area. Carbon fixation, soil conservation, habitat quality and water purification service show different degrees of fluctuation within the three metropolitan areas, with the greatest fluctuation in C-Z-T metropolitan area. At different spatial scales, LUCC and its telecoupling effect have a significantly negative effect on ES, and with the expansion of the scale, the negative effect will be strengthened gradually, while the influences of LUCC and its telecoupling effects on ES differ significantly in different levels of urban agglomerations. This paper can provide a reference for the implementation of differentiated land use and ecological protection policies for different levels of urban agglomerations.

Spatial spillovers and world energy intensity convergence

This paper studies the influence of spatial spillovers on energy intensity growth and convergence across 153 countries from 1999 to 2018. The inclusion of spatially lagged variables allows the study of the influence of trade, technological diffusion and policy mimicry on these two processes. Contrary to previous literature, we consider different spatial specifications and weight matrices to test whether spatial spillovers are exogenous or endogenous, as well as their range of influence. Our study finds the spatial lag of X model (SLX), with local spatial weight matrices, fits better to the data. Convergence is found for the world as a whole, conditional to the spatial distribution of countries and several long-term economic characteristics. In this sense, clusters of nearby countries present higher rates of convergence. Besides, domestic capital accumulation, total factor productivity growth and renewable energy consumption significantly determine the characteristics of clubs of convergence. Furthermore, spatial spillovers associated to capital accumulation, population growth and renewable energy consumption also contribute to the definition of these clubs.

The Impact of Green Finance on High-Quality Economic Development in China: Vertical Fiscal Imbalance as the Moderating Effect

The vigorous development of green finance has become a national strategy in China. Green finance is gradually becoming a key driver of high-quality economic development and a key area of concern for China’s economy and ecological environment. Based on the panel data of 30 Chinese provinces from 2001 to 2019, we analyzed the impact and mechanism of vertical fiscal imbalance (VFI) and green finance (GF) on high-quality economic development (HQD) and then used the fixed-effect model and spatial Durbin model for empirical testing. We found that GF can significantly contribute to HQD, but VFI has a negative moderating effect in contrast to the positive effect of GF on HQD. This negative moderating effect is strongest in the central region. According to the analysis of the spatial econometric model based on geography, economy, and the nested spatial weight matrix, we found that the local GF has a negative spatial transmission effect on the HQD of other regions. Therefore, it is recommended that the coordinated development of green finance among regions be promoted, while affairs and expenditure responsibilities be reasonably distributed between the central and local governments to drive HQD effectively.

A vision subsampling probsparse self-attention-based informer for bearing remaining useful life prediction

Bearing remaining useful life (RUL) prediction has always been a central topic in the industry field, the aim of which is to optimize system safety and sustainability. The validity of prediction models and the accuracy of prediction results are affected by mid-term singularities and terminal mutations, under time-domain bearing vibration information. In this paper, a network structure-cascaded dilated convolution vision informer (CDC-Vii) is put forward to precisely forecast the RUL of bearings, which uses the time–frequency fault features as input. CDC-Vii breaks the limitation of the original Informer, which is only sensitive to time-series information. An adaptive fault frequency band selection algorithm is proposed, which can reduce training time while utilizing rich time–frequency information. Based on the Informer architecture, the attention mechanism is improved to form vision subsampling probsparse self-attention (VSPS). VSPS can precisely assign spatial attention weights and reduce computational complexity. At the same time, a truncated relative position encoding technique is proposed to strengthen the position dependence between attention information. Moreover, cascaded dilated convolution enhances the image contrast of faulty frequency bands while enlarging the use of the receptive field. Experiments on two extensively utilized bearing datasets reveal that CDC-Vii surpasses the advanced RUL prediction models.

Forecasting per Capita Energy Consumption in China Using a Spatial Discrete Grey Prediction Model

To overcome the limitations of the present grey models in spatial data analysis, a spatial weight matrix is incorporated into the grey discrete model to create the SDGM(1,1,m) model, and the L1-SDGM(1,1,m) model is proposed, considering the time lag effect to realize the simultaneous forecasting of spatial data. The validation of the SDGM(1,1,m) and L1-SDGM(1,1,m) models is achieved, and finally, the per capita energy consumption levels (PCECs) of 30 provinces in China from 2020 to 2025 is predicted using SDGM(1,1,m) with a metabolic mechanism. We draw the following conclusions. First, the SDGM(1,1,m) and L1-SDGM(1,1,m) models established in this paper are reasonable and improve forecasting accuracy while supporting interactive regional forecasting. Second, although SDGM(1,1,m) resembles the DGM(1,n) model, their modeling conditions and targets are different. Third, the SDGM(1,1,m) and L1-SDGM(1,1,m) models can be used to effectively analyze the spatial spillover effects within the selected modeling interval while achieving accurate predictions; notably, from 2010 to 2017, the PCECs of Inner Mongolia and Qinghai were most affected by spatial factors, while the PCECs of Jilin, Jiangxi, and other provinces were influenced little by spatial factors. Fourth, predictions indicate that the PCECs of most Chinese provinces will increase under the current grey conditions, while the PCECs of provinces such as Beijing are expected to decrease.

A Multi-Feature Fusion Model Based on Denoising Convolutional Neural Network and Attention Mechanism for Image Classification

Spatial location features extracted by denoising convolutional neural network. At this time, an attention mechanism is introduced into denoising convolutional neural network. The dual attention model of local area is presented from two dimensions of channel and space—channel attention mechanism weights channel and spatial attention mechanism weights location. A variety of machine learning methods are used to classify and train different features. Multi-semantic features and heterogeneous features are fused by adaptive weighted fusion algorithm. Finally, the data sets Cifar-10, STL-10, Cifar-100 and GHIM-1OK are verified on the proposed method. Compared with a single semantic feature, the accuracy is improved by 10%-15%. Compared with several advanced algorithms, the performance has a significant advantage, which proves the complementarity of heterogeneous features and multi-network semantic features and the effectiveness of the adaptive weighted fusion algorithm.