High-performance solutions of geographically weighted regression in R

Lu, Binbin; Hu, Yigong; Murakami, Daisuke; Brunsdon, Chris; Comber, Alexis; Charlton, Martin; Harris, Paul

doi:10.1080/10095020.2022.2064244

Cited by 21 publications

(6 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Notably, an R shiny package, namely GeoWeightedModel [40] has been developed to specifically provide a graphical user interface for GW functionalities in GWmodel. Secondly, memory and computational limits explicitly exist in R, although high-performance solutions have been expediently developed with multi-core or compute unified device architecture configurations [41]. This natural limitation of R tend to lead the efficiency of high-performance solutions to be compromised.…”

Section: Motivation and Significancementioning

confidence: 99%

See 1 more Smart Citation

GWmodelS: A software for geographically weighted models

Lu¹,

Hu²,

Yang³

et al. 2023

SoftwareX

View full text Add to dashboard Cite

Section: Motivation and Significancementioning

confidence: 99%

“…(4) Computational options: In practice, stand-alone computing is fine for most of the GW models, particular when the sample size is small (e.g. less than 3000) [41]. With concerning large scale data, high-performance options, including multi-core parallel and CUDA techniques are also seamlessly integrated in this software.…”

Section: Software Functionalitiesmentioning

confidence: 99%

GWmodelS: A software for geographically weighted models

Lu¹,

Hu²,

Yang³

et al. 2023

SoftwareX

View full text Add to dashboard Cite

“…As an extension of the traditional least squares regression method, GWR was first proposed by Brunsdon (Brunsdon et al, 1998), which described the dynamic relationship between the causal variables and explanatory variables in the geographic space, rather than the single regression that took the study area as a whole (Lu et al, 2022). Therefore, it has been widely applied in the study of complex scale-dependent characteristics between meteorological elements and topographic elements in high-altitude mountainous areas (Chu, 2012;Kumari et al, 2017).…”

Section: Gwr and Mgwr Regression Modelsmentioning

confidence: 99%

Construction of high-resolution precipitation dataset and its implication to drought over the Tianshan Mountains, China

He²,

Li³

et al. 2023

Front. Earth Sci.

View full text Add to dashboard Cite

Understanding the drought characteristics of mountainous areas in northwest China with sparse rainfall stations requires high precision, as well as high-resolution precipitation data. Considering the spatial relationship of precipitation and environmental factors, this study downscales Global Precipitation Measurement (GPM) and Multi-Source Weighted-Ensemble Precipitation (MSWEP) based on the geographically weighted regression (GWR) and multi-scale geographically weighted regression (MGWR) models integrated with interpolation. A high-resolution (1 km×1 km) precipitation dataset during 1979–2020 is reconstructed in the Tianshan Mountains, and the drought characteristics are analyzed by using the optimal dataset. The results show that: 1) Compared with GWR, MGWR model has higher downscaling accuracy; 2) The optimal MSWEP downscaling dataset (CC = 0.93, |BIAS| = 0.48%) compared to GPM (CC = 0.81, |BIAS| = 1.87%) is closer to the observed precipitation; 3) In the past 40 years, 71% and 9% of the Tianshan Mountains show significant wetting and drying trends respectively, and 16 drought events are identified. 4) The West subregion of the Tianshan Mountains is characterized by low frequency, long duration and high severity of drought events. The characteristics of the East are opposite to those of the West. Occasional extreme drought events occur in the North and South. This paper provides data support and method reference for the study of water-vapor balance and regional ecohydrological process in the arid area of Northwest China.

show abstract

“…The variable is significant at a high level (i.e., t-statistics are very high or p-value is very low) [40,41]. The results of the models in the table compared the precision between GWR and OLS (ordinary least square) models to visualize the difference in the accuracy of the models [42]. The alternative model proposes four alternative models: Y%ov1, Y%ov2, Y%ov3, and Y%ov4, as shown in Table 5.…”

Section: Optimal Gwr Model For Predicted With Liver Fluke (Opisthorch...mentioning

confidence: 99%

Spatial Analysis of Influence Factors Associated With Liver Fluke (Opisthorchis viverrini) Infection in Small Sub-Watershed Using GWR Modeling

Pumhirunroj,

Littidej,

Boonmars

et al. 2023

Preprint

View full text Add to dashboard Cite

Infection of liver flukes (Opisthorchis viverrini) is partly due to their suitability for habitats in sub-basin areas, which causes the intermediate host to remain in the watershed system in all seasons. Spatial monitoring of fluke infection at the small -basin analysis scale is important because this can enable analysis at the level of the spatial factors involved and influencing infections. A geographic weighted regression model was developed to analyze the spatial characteristics of liver fluke infection, aiming to 1. analyze the spatial factors associated with human liver fluke infection according to sub-basin boundaries and 2. generate an alternative model for enhancing the effectiveness of preventive public health management to reduce the risk of liver fluke infection in humans. The number of infected persons was obtained from local authorities and converted into a percentage of infected people and generated as raster data with a heat map so that the data were continuous and defined as dependent variables. The independent set consisted of nine variables, both vector and raster data, that correlated the location with the village location of an infected person. The results showed that the variables X5stream, X7ndmi, and X9savi were statistically significantly correlated to the percentage of infected people, with the t-stat and p-value being (-2.068, 1.875, and -2.661) and (0.048, 0.034, and 0.021), respectively. The GWR model was able to increase accuracy more than the comparable models such as OLS, in all tests of the four alternative models, with an accuracy increase in R2 of 7.69% (0.576 to 0.624). This study confirms that the development of spatial models with GWR models can screen for factors associated with liver fluke infection at the level of small spatial units such as sub-basins.

show abstract

High-performance solutions of geographically weighted regression in R

Cited by 21 publications

References 44 publications

GWmodelS: A software for geographically weighted models

GWmodelS: A software for geographically weighted models

Construction of high-resolution precipitation dataset and its implication to drought over the Tianshan Mountains, China

Spatial Analysis of Influence Factors Associated With Liver Fluke (Opisthorchis viverrini) Infection in Small Sub-Watershed Using GWR Modeling

Contact Info

Product

Resources

About