Table 3 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Context in source publication
Similar publications
Over the recent years, spatial statistics are gaining ground against traditional statistics for the interpretation of spatial phenomena, with respect to taking into account space impact. Geographically Weighted Regression (GWR) is a contemporary spatial statistics method, a modern approach of the classic regression model in which the degree of rela...
The recent increase in energy consumption worldwide has accelerated global warming. Thus, developed countries are aiming to reduce energy consumption in cities and promote eco-friendly policies. Buildings account for most of the energy used in a city. Therefore, it is necessary to identify the factors that affect electrical energy consumption in ur...
The gold (Au) geochemical anomaly is an important indicator of gold mineralization. While the traditional field geochemical exploration method is time-consuming and expensive, the hyperspectral remote sensing technique serves as a robust technique for the delineation and mapping of hydrothermally altered and weathered mineral deposits. Nonetheless,...
In India, cotton yield estimates are done using crop cutting experiments (CCEs) conducted within the framework of the general crop estimation surveys (GCES) methodology. In recent times, for obtaining reliable estimates at levels lower than the district, the number of CCEs has increased in comparison to the existing setup of GCES. This puts an addi...
There is a paucity of studies investigating the spatial pattern and factors associated with the prevalence of tuberculosis (TB) in Java. This study aimed to identify spatial autocorrelation, clusters, and factors associated with TB prevalence in Java using district-or city-level data. This was an ecological study using data from 118 districts or ci...
Citations
... Spatial dependency (or autocorrelation) of residuals calls for using a Geographically Weighted Regression (GWR) to further examine how the statistical relationship between dependent and independent variables varies over space (Nkeki & Osirike, 2013;Pimpler, 2017). GWR is a localized version of OLS regression that was proposed in the geography field in 1996 to show how regression model relationships vary over space (Wheeler, 2014). ...
Criminal activity is unevenly spread over space, necessitating the intervention of local residents to combat the threat. This study seeks to examine the overall patterning of the effect of socioeconomic/demographic indicators and crime rates on indigenous police force numbers in Nigeria. Both the rates of indigenous police number and crime were studied using spatial and quantitative methodologies, which was based on data from National Bureau of Statistic (2017, 2018) as well as the educational index, unemployment rate, and % male. Crime rates and socioeconomic/demographic factors were investigated using Ordinary Least Squares (OLS) linear regression in ArcGIS 10.5 environment to see how they connect to the need for more indigenous police officers at the state level. To further comprehend the spatially changing associations between the rate of indigenous police number and the independent variables included in the OLS model, Geographically Weighted Regression (GWR) was utilized. Findings revealed that the states of Adamawa, Akwa-Ibom, Benue, Edo, and Kogi had the highest rates of indigenous police officers, whereas Anambra, Bauchi, Federal Capital Territory (FCT), Jigawa, Kano, Lagos, Oyo, and Sokoto had the lowest indigenous police numbers. The study found that the states of Delta, Ebonyi, Lagos and the FCT recorded high rate of theft/stealing due to lower rate of indigenous police number and high population density. According to the study, the rate of indigenous police officers diminishes as population density rises. Also, the rate of indigenous police officers was highly connected with rates of robbery and theft/stealing. This study used GWR to highlight the spatially varying relationships between the socioeconomic/demographic indicators and various crime rates on rate of indigenous police officers in Nigeria, which will help Federal Government of Nigeria, decide which appropriate measures are needed for concentrating on states with high crime rates that need more local policing. The study emphasizes the need for safe states being established not just by law enforcement but by a wide range of social and economic sectors and services, as well as integrating more indigenous people in state policing.
... Spatial dependency (or autocorrelation) of residuals calls for using a Geographically Weighted Regression (GWR) to further examine how the statistical relationship between dependent and independent variables varies over space (Nkeki & Osirike, 2013;Pimpler, 2017). GWR is a localized version of OLS regression that was proposed in the geography field in 1996 to show how regression model relationships vary over space (Wheeler, 2014). ...
... Where Y is the dependent variable (BLB prevalence), the betas β0 to βn represent the consequent number of the coefficients of predictors while 1 n X to X depicts the corresponding number of predictors and ε is error of residuals. Ordinary least square ANOVA contain different statistical tests which includes Joint F-statistics, Koenker statistics, Wald statistics and Jarque and Bera statistics which define the explanatory variables are significant to independent value or not (Nkeki and Osirike, 2013;Ahmad et al., 2021). Environmental factors that were used as explanatory variables were: Humidity, Surface pressure data was taken from NASA Power Data Access Viewer • Soil pH, and Soil Organic Carbon data taken from soil grids • Elevation by Diva GIS Data was interpolated by kriging method; it is a process that includes data from known nearer points to estimate Where, θ is maximum likelihood estimator and n I θ is expected fisher information. ...
... which independent variable has the greatest influence in a particular region? (Nkeki and Osirike, 2013). ...
Xanthomonas oryzae pv. oryzae (Xoo) causes bacterial leaf blight that is a major threat to rice production. Crop losses in extreme situations can reach up to75%, and millions of hectares of rice are affected each year. Management of the disease required information about the spatial distribution of BLB incidence, severity, and prevalence. In this study, major rice-growing areas of Pakistan were surveyed during 2018-2019 for disease occurrence, and thematic maps were developed using geographic information system (GIS). Results showed that Narowal district had highest percentage of disease incidence (54-69%), severity (42-44%), and prevalence (72-90%) meanwhile Jhung district had the lowest incidence (21-23%), severity (18-22%), and prevalence (45-54%). To understand the environmental factors contributing to this major rice disease, the research analyze, the spatial relationships between BLB prevalence and environmental variables. Those variables include relative humidity (RH), atmospheric pressure (A.P), minimum temperature, soil organic carbon, soil pH, and elevation, which were evaluated by using GIS-based Ordinary Least Square (OLS) spatial model. The fitted model had a coefficient of determination (R2) of 65 percent explanatory power of disease development. All environmental variables showed a general trend of positive correlation between BLB prevalence and environmental variables. The results show the potential for disease management and prediction using environmental variable and assessment.
... cases, an initial classification of data was conducted. COVID-19 cases and property transactions data in this study were classified using the natural breaks method, which is based on the Jenks Natural Breaks algorithm (Nkeki and Osirike, 2013). The Jenks Natural Breaks algorithm method produced class breaks that identified the best group of similar values and maximized the differences between the classes of the registered COVID-19 cases and property transactions. ...
The real estate sector in Uganda has been substantially impacted by the onset of COVID-19 in this country. Studies conducted worldwide have indicated that, pandemics affect property market activities differently. Additionally, the effect of pandemics on property market activity varies from one place to another. Studies conducted in Uganda, however, have not captured how the effect of COVID-19 on property market activities varies from one place to another. This study therefore explored the spatial variability of the effect of COVID-19 on property market activities in Kampala district, Uganda. The study took advantage of the spatial statistical analytical models advocated by GIS (Getis-Ord Gi*, OLS, GWPR) and a unique dataset of property transactions registered by the Ministry of Lands, Housing and Urban Development (MLHUD) during the outbreak of the deadly disease. Whereas the study observed high volumes of property transactions registered in the residential outskirts of the city, low volumes were observed in the Central Business District (CBD) and the low-income areas of the eastern and western parts of the district. On the other hand, the local model approach of GWPR exposed the substantial effects of COVID-19 on property market activities that varied from -39% to 10%. It was further established that COVID-19 generated negative effects in areas with low and high prices of land per acre, to the extent of increasing as the prices dropped or increased. On the contrary, a positive effect was realized in the residential outskirts of the city where prices of land per acre were moderate. Work from home, land parcel size as well as the composition of the population, proved to be the main drivers of the changes in property market transactions (activity). The findings of the study underpin the earlier postulations of various researchers that pandemics affect property market activity. However, the effects of the pandemics vary from one pandemic to another and from one place to another.
... Weighted Regression (GWR) to further examine how the statistical relationship between dependent and independent variables varies over space (Nkeki & Osirike, 2013;Pimpler, 2017). GWR is a localized version of OLS regression that was proposed in the geography field in 1996 to show how regression model relationships vary over space (Wheeler, 2014). ...
Criminal activity is unevenly spread over space, necessitating the intervention of local
residents to combat the threat. This study seeks to examine the overall patterning of the effect
of socioeconomic/demographic indicators and crime rates on indigenous police force
numbers in Nigeria. Both the rates of indigenous police number and crime were studied using
spatial and quantitative methodologies in this study, which was based on data from National
Bureau of Statistic (2017, 2018) as well as the educational index, unemployment rate, and %
male. Crime rates and socioeconomic/demographic factors were investigated using Ordinary
Least Squares (OLS) linear regression in ArcGIS 10.5 environment to see how they connect
to the need for more indigenous police officers at the state level. To further comprehend the
spatially changing associations between the rate of indigenous police number and the
independent variables included in the OLS model, Geographically Weighted Regression
(GWR) was utilized. Findings revealed that the states of Adamawa, Akwa-Ibom, Benue, Edo,
and Kogi had the highest indigenous police numbers, whereas Anambra, Bauchi, Federal
Capital Territory (FCT), Jigawa, Kano, Lagos, Oyo, and Sokoto had the lowest indigenous
police numbers. The study found that the states of Delta, Ebonyi, Lagos and the FCT
recorded high rate of theft/stealing due to lower rate of indigenous police number and high
population density. According to the study, the number of indigenous police officers
diminishes as population density rises. Also, the rate of indigenous police officers was highly
connected with theft/stealing. GWR's findings explained variations in the indigenous police
number and identified spatially non-stationary correlations. The study emphasizes the need
for safe states being established not just by law enforcement but by a wide range of social and
economic sectors and services, as well as integrating more indigenous people in state
policing.
... Studies have proven that incorporating GWR into ArcGIS package produces betterquality quality of possible results. For instance, Nkeki, and Osirike [35] shows that the GIS-based GWR can be used for spatial presentation of the parameter estimates and coefficient of determination as regard all variables in a raster surface as well as vector map. ...
Geographic Information system (GIS) is a powerful computer-based tool used in building spatially explicit models for comprehending real-world processes. Consequently, it has attracted extensive research efforts over the past half-century across the globe. Also, spatial analysis is a significant area of application within geographic information science or a computer-enhanced geographic data environment. The reason is that virtually all the occurrences we encounter daily assume a spatial context. GIS is suitable and effective for providing the requisite solutions to various problems related to these manifestations including representing, analyzing, and knowing their spatial dimensions. Thus, this review focuses on the relationship between GIS and the evolution of spatial analysis. It offers an in-depth discussion of the evolutionary theory of spatial analysis with a specific emphasis on quantitative geography, regional science, spatial statistics, and computational geometry. Summarily, GIS is highly effective in handling spatial analysis because of its ability to manage both planar and attribute data in an integrated manner.
... According to [53] [54], improvements in Geographical Information Systems (GIS) technology offers numerous opportunities for epidemiologists to study the spatial distribution of diseases and also understand the relationships between socio-economic, behavioral and environmental factors and the occurrence of diseases. For instance, GIS techniques have been employed in the investigation and monitoring of vector-borne diseases [55] [56], health education [57] waterborne diseases [58] [59], as well as in environmental health [60] [61]. Geospatial Information System is a software that links geographic information with descriptive information [62]. ...
... Dziauddin, 2019;Mulley, 2014). However, global statistical methods often assume a homogeny (stationary) relationship between the dependent and predictor variables over space, which could lead to misleading results whenever applied to a spatial dataset (Nkeki & Osirike, 2013;So, Tse, & Ganesan, 1997). In 2002, to reinforce this weakness, a more refined alternative -e.g. ...
... It is imperative to check the performance of the global model before proceeding further. The performance of the global model is tested through residual (Nadi & Murad, 2019;Nkeki & Osirike, 2013). Non-clustered residuals indicate that the global model i.e. ordinary least square (OLS) has good performance. ...
... However, the diseases and related affecting factors may show obvious spatial heterogeneity in the real world. So, global regression methods may provide inaccurate predictions (Nkeki and Osirike, 2013). In contrast, geographically weighted regression (GWR) is a local regression method that detects spatial heterogeneity of the factors (Chan et al., 2014;Kim et al., 2019;Koh et al., 2020). ...
Numerous methods have been implemented to evaluate the relationship between environmental factors and respiratory mortality. However, the previous epidemiological studies seldom considered the spatial and temporal variation of the independent variables. The present study aims to detect the relations between respiratory mortality and related affecting factors across Xi'an during 2014–2016 based on a novel geographically and temporally weighted regression model (GTWR). Meanwhile, the ordinary least square (OLS) and the geographically weighted regression (GWR) models were developed for cross-comparison. Additionally, the spatial autocorrelation and Hot Spot analysis methods were conducted to detect the spatiotemporal dynamic of respiratory mortality. Some important outcomes were obtained. Socioeconomic and environmental determinants represented significant effects on respiratory diseases. The respiratory mortality exhibited an obvious spatial correlation feature, and the respiratory diseases tend to occur in winter and rural areas of the study area. The GTWR model outperformed OLS and GWR for determining the relations between respiratory mortality and socioeconomic as well as environmental determinants. The influence degree of anthropic factors on COPD mortality was higher than natural factors, and the effects of independent variables on COPD varied timely and locally. The results can supply a scientific basis for respiratory disease controlling and health facilities planning.
... To investigate the urban energy demand modeling, the best combinations of explainers are applied to the OLS and GWR analysis for the next step. OLS is a commonly used global modeling technique to evaluate relationships between two or more variables without spatial variation and relationships within geographic entitles (Nkeki & Osirike, 2013). GWR is a location-dependent method to investigate the spatial relationship between explanatory variables (explainers) and a dependent variable, such as energy demand (Zhou & Wang, 2010). ...
Considering climate change and energy resource depletion under rapid urbanization trends in the urban environment, the relation between land-use, climate change, and urban energy demand is gaining attention. However, a limited number of studies are focusing on the effect of microclimate change, and more specifically, temperature change on energy demand at an urban scale. This study includes empirical spatial and temporal modeling to identify how urban morphology indicators (UMIs), land surface temperature (LST), and neighboring land-use compositions affect urban energy demand using an extensive data set for the case study of Eindhoven, the Netherlands. For this purpose, the ordinary least square regression (OLS) and geographically weighted regression (GWR) models are employed. The results show, there is a significant spatial relation between UMIs, neighboring land-use compositions, and urban energy demand. Furthermore, the impact of dwelling types on urban energy demand is discussed. The results can be applied to sustainable urban planning targeting energy reduction, climate adaptation, and help local authorities for implementing energy management strategies.
