Spatial self-confounding: smoothness-related estimation bias in spatial regression models

An empirical evaluation of spatial regression models

Spatial regression models are developed as a complementary alternative to second-order polynomial response surfaces in the context of process optimization. These models provide estimates of design variable effects and smooth, data-faithful approximations to the unknown response function over the design space. The predicted response surfaces are driven by the covariance structures of the models. Several structures, isotropic and anisotropic, are considered and connections with thin plate splines are reviewed. Estimation of covariance parameters is achieved via maximum likelihood and residual maximum likelihood. A feature of the spatial regression approach is the visually appealing graphical summaries that are produced. These allow rapid and intuitive identification of process windows on the design space for which the response achieves target performance. Relevant design issues are briefly discussed and spatial designs, such as the packing designs available in Gosset, are suggested as a suitable design complement. The spatial regression models also perform well with no global design, for example with data obtained from series of designs on the same space of design variables. The approach is illustrated with an example involving the optimization of components in a DNA amplification assay. A Monte Carlo comparison of the spatial models with both thin plate splines and second-order polynomial response surfaces for a scenario motivated by the example is also given. This shows superior performance of the spatial models to the second-order polynomials with respect to both prediction over the complete design space and for cross-validation prediction error in the region of the optimum. An anisotropic spatial regression model performs best for a high noise case and both this model and the thin plate spline for a low noise case. Spatial regression is recommended for construction of response surfaces in all process optimization applications.

In the present study, we investigate different factors related to demography, land use, transportation and height that influence the generation of the PM2.5 contaminant due to woodburning in the conurbation of Temuco and Padre Las Casas, Chile. First, a global regression model was implemented to explore the relationships between the variables, and subsequently, spatial regression models were implemented (geographically weighted regression and spatial error model). Results suggest that height and infrastructure density have a strong association with the PM2.5 contaminant using the global regression and spatial error models. However, results indicate that residential density, distance to high vehicle congestion, and length of local roads within a vicinity of 100 m are significant when employing the geographically weighted regression. In addition, the latter type of regression model presents the best performance when compared to the other two models (global regression and spatial error). These results may help authorities in the allocation of resources to improve air quality in the studied conurbation. Future research should include other types of spatial regression models and other relevant variables that may influence the generation of PM2.5 due to woodburning.

Energy monitoring using statistical process control (SPC) methods makes it more straightforward to identify patterns and trends to decrease energy consumption more effectively. The literature review of energy consumption monitoring with SPC techniques generally focuses on the temporal aspect of variation. However, due to the spatial nature of energy data, enhancing these methods to incorporate temporal and spatial aspects would improve the accuracy of the diagnostic information, underscoring simultaneous detection of the time and location of changes. Thus, the main novelty of this work is the spatial modeling and spatiotemporal monitoring of electricity consumption. For this purpose, the study used actual electricity consumption data from eight western cities of Mazandaran province in the north of Iran for spatial modeling using spatial regression models and a geographically weighted regression (GWR) model. The prediction performance evaluation of spatial models showed GWR as an appropriate model, whose coefficients were monitored through a generalized likelihood ratio (GLR) chart in phase II. The GLR chart detected two changes in consumption, and its performance was confirmed based on the statements from electricity experts relying on meteorological information and floating population data. Furthermore, the performance of the GLR chart was evaluated using out-of-control average run length (ARL1) across three different scenarios. The findings indicate that the GLR chart can effectively detect any sizes of shifts (δ), ranging from 5% to 100% of the model's parameter value. Additionally, with larger values of δ, the ARL1 decreases, resulting in faster detection of changes in the model.

Modeling the impact of land use/land cover (LULC) factors on diurnal and nocturnal Urban Heat Island (UHI) intensities using spatial regression models

Tree height–diameter relationships are usually studied using linear or nonlinear models, but exogenous variables, especially spatially autocorrelated and dependent variables of tree diameter or height, are not often considered in height–diameter modeling. Three types of spatial regression models — spatial lag model, spatial error model, and spatial Durbin process model — are explored in this study. The height–diameter relationships are modeled using the spatial regression models to investigate the effects of spatial dependence and spatial autocorrelation and the roles of the exogenous variables generated by neighboring trees. Case study 1 shows that the spatial lag model should be used to analyze height–diameter relationships, in which heights of neighboring trees, which are exogenous variables, and the endogenous variable DBH significantly affect height growth. Case study 2 shows that the spatial error model performs better than other models, and that height growth is not only affected by its endogenous variable diameter but also by unobserved variables that vary spatially and result in residual spatial autocorrelation. Spatial regression models are an approach to height–diameter modeling that provide insight into how the endogenous variable diameter, the exogenous variables height and (or) diameter of neighboring trees, and locally varied but unobserved environmental or ecological variables contribute to height growth.

Pedestrian safety has recently been considered as one of the most serious issues in the research of traffic safety. This study aims at analyzing the spatial correlation between the frequency of pedestrian crashes and various predictor variables based on open source point-of-interest (POI) data which can provide specific land use features and user characteristics. Spatial regression models were developed at Traffic Analysis Zone (TAZ) level using 10,333 pedestrian crash records within the Fifth Ring of Beijing in 2015. Several spatial econometrics approaches were used to examine the spatial autocorrelation in crash count per TAZ, and the spatial heterogeneity was investigated by a geographically weighted regression model. The results showed that spatial error model performed better than other two spatial models and a traditional ordinary least squares model. Specifically, bus stops, hospitals, pharmacies, restaurants, and office buildings had positive impacts on pedestrian crashes, while hotels were negatively associated with the occurrence of pedestrian crashes. In addition, it was proven that there was a significant sign of localization effects for different POIs. Depending on these findings, lots of recommendations and countermeasures can be proposed to better improve the traffic safety for pedestrians.

The purpose of this study is to examine the impact of the urban environment on surface temperature on the Seoul in order to find ways to reduce urban heat islands. There are four factors that affect surface temperature: land use and land cover, transportation, Atmospheric environment, and population. Moran s I test of surface temperature showed strong spatial autocorrelation, which was analyzed using the spatial regression model, and the spatial lag model (SLM) was the most suitable among the spatial regression models, so it was selected as the final model. According to the analysis, the residential area ratio, the bus stop density, and the intersection density were positively correlated with the surface temperature, while wind speed and NDVI were negatively correlated. In the analysis process, a more specific analysis could be performed by setting the 100 m x 100 m grid as the spatial unit of the analysis, and a more statistically significant result could be obtained by using a spatial regression model. This study is meaningful in that it is the basis for presenting direction in urban heat island mitigation and adaptation policies.

We examine the determinants of Airbnb prices in 10 major EU cities, focusing on the role of location. The results confirm that attributes related to size, quality, and location are all significant drivers of Airbnb rates. Novel indices based on TripAdvisor data are used to measure the attractiveness of neighbourhoods, and the results show a more robust impact on price than standard location variables based on selected points of interest. The analysis confirms that Airbnb prices are spatially dependent, requiring the implementation of spatial regression models. Following recent studies on spatial econometrics, we examine various spatial models, including specifications with multiple sources of spatial dependence. The results show significant differences between the coefficients estimated with OLS and the various spatial models, especially in the case of location-specific variables. As well as having managerial and policy implications, our study contributes to the hedonic price literature by providing a methodological guide on spatial regression models. • Airbnb prices are spatially dependent. • Attractiveness of neighbourhoods can be measured with TripAdvisor data. • Accessibility of tourist attractions is a major price driver. • Multi-source spatial models provide best fit to data.

The purpose of this study was to assess the spatial distribution of malaria prevalence rates among selected rural part of woredas in SNNPR, Ethiopia. This work is based on data available from the 2011 malaria indicator survey (MIS 2011) of Ethiopian Public Health Institution. ESDA, Spatial regression model and Bayesian Spatial analysis were employed for data analysis. From ESDA, we found positive spatial autocorrelation in malaria prevalence rate. Relying on specification diagnostics and measures of fit; Spatial lag model was found to be the best model for modeling malaria prevalence rate data. The relationship between malaria prevalence and its risk factors was assessed using spatial models. The spatial models also showed an increase of malaria prevalence with a number of factors. From results, increase in the proportion of households sprayed in 12 months and the average altitude in the woreda estimated to decrease the average malaria prevalence. The result also demonstrated that increase in the House hold size of the district, proportion of households having access to piped water, proportion of households having access to radio, proportion of households having access to radio and Main construction material of the room’s wall are estimated to raise the average malaria prevalence rate. Finally, the study concluded that malaria is spatially clustered in space and the risk factors exhibit effect on the malaria prevalence in the study area. Based on the results of the study, We recommend for policy makers on the way to reduce malaria prevalence in the rural part of woreda of SNNPR using spatial information.

Development of spatially varying groundwater-drawdown functions for land subsidence estimation

Typically, a Poisson model is assumed for count data. In many cases, there are many zeros in the dependent variable, thus the mean is not equal to the variance value of the dependent variable. Therefore, Poisson model is not suitable anymore for this kind of data because of too many zeros. Thus, we suggest using a hurdle‐generalized Poisson regression model. Furthermore, the response variable in such cases is censored for some values because of some big values. A censored hurdle‐generalized Poisson regression model is introduced on count data with many zeros in this paper. The estimation of regression parameters using the maximum likelihood method is discussed and the goodness‐of‐fit for the regression model is examined. An example and a simulation will be used to illustrate the effects of right censoring on the parameter estimation and their standard errors.

Spatial regression model is used to determine the relationship between the dependent and independent variables with spatial influence. In case only independent variables are affected, Spatial Cross Regressive (SCR) Model is formed. Spatial Autoregressive (SAR) occurs when the dependent variables are affected, while Spatial Durbin Model (SDM) exists when both variables exhibit effects. The inaccuracy of the spatial regression model can be caused by outlier observations. Removing outliers in the analysis changes the spatial effects composition on data. However, using robust spatial regression is one way of overcoming the outliers in the model. Moreover, the typical parameter coefficients, which are robust against the outliers, are estimated using M-estimator. The research develops the life expectancy model in Central Java Province through Robust-SCR, Robust-SAR, and Robust-SDM to reduce spatial outliers' effect. The model is developed based on educational, health and economic factors. According to the results, M-estimator accommodates the outliers’ existence in the spatial regression model. This is indicated by an increase in R2 value and a decrease in MSE caused by the change in the estimating coefficient parameters. In this case, Robust-SDM is the best model since it has the biggest R2 value and the smallest MSE.

While spatial data analysis has received increasing attention in demographic studies, it remains a difficult subject to learn for practitioners due to its complexity and various unresolved issues. Here we give a practical guide to spatial demographic analysis, with a focus on the use of spatial regression models. We first summarize spatially explicit and implicit theories of population dynamics. We then describe basic concepts in exploratory spatial data analysis and spatial regression modeling through an illustration of population change in the 1990s at the minor civil division level in the state of Wisconsin. We also review spatial regression models including spatial lag models, spatial error models, and spatial autoregressive moving average models and use these models for analyzing the data example. We finally suggest opportunities and directions for future research on spatial demographic theories and practice.

پتانسیل تجاری اتحادیه اکو: کاربردی از رگرسیون فضایی پویا با در نظرگرفتن اثرات سرریز فضایی