Application of Geographically Weighted Regression Technique in Spatial Analysis of Fatal and Injury Crashes

Abstract

Generalized Linear Models (GLMs) are the most widely used models utilized in crash prediction studies. These models illustrate the relationships between the dependent and explanatory variables by estimating fixed global estimates. Since the crash occurrences are often spatially this is due to the capability of GWGLMs models in capturing the spatial heterogeneity of crashes.

Situation analysis of healthcare service delivery using geographically weighted regression: (A case study of Sironko District, Eastern Uganda)

“…5km kernel distance was arrived at after iteration with increasing bandwidth distance from 1km to 10km, and checking the regression diagnostics to ensure that the model passes and performs best. Three assumptions suggested by Chi et al (2016), Pirdavani et al (2014) and Sinaga et al (2016) The variance of error terms should be constant across observations (also known as homoskedasticity), and [3] The error terms are not auto-correlated and any one residual is not correlated with any other residual in the study area. The main check was the R-squared (ranging from 0.0 (0%) to 1.0(100%)) with the higher value indicating the better model in terms of performance.…”

“…Developing an efficient framework for the regression model requires that there are set assumptions, the Kernel Weighting Function is established, the choice of the bandwidth is made and most importantly, model diagnostic tests are performed to ensure that the parameter estimates of the resultant regression model are not biased (Pirdavani et al, 2014;Sinaga et al, 2016). According to Rhee et al (2016), generalized linear models (GLMs) are conventional linear regression models for a continuous response variable given a set of continuous and/or categorical predictors.…”

“…Matthews et al (2012) argued that GWR can potentially make significant contributions to health research, such as allowing researchers to better understand the etiology and spatial processes, offering informative results beyond global models to facilitate place-specific health policy formation, and enabling scholars to explore questions that cannot be answered with traditional (global) analytical models. GWR was preferred to Ordinary Least Squares (OLS) for regression as the latter produces a global model which doesn't accommodate spatial non-stationarity of georeferenced data, (Lin et al, 2011 andPirdavani et al, 2014). OPD2015 was used as the dependent variable for the regression analysis performed in ArcMap 10.2.2 using the GWR tool.…”

Road Safety Risk Evaluation Using GIS-Based Data Envelopment Analysis—Artificial Neural Networks Approach

“…In Belgium, through the use of GIS and point pattern techniques, mapping road-accident black zones has been conducted within urban agglomerations [41]. GIS has also been used to explore the spatial variations in relationship between Number of Crashes and other explanatory variables of 2200 Traffic Analysis Zones (TAZs) in the study area, Flanders, Belgium [42,43]. GIS was used for modelling crash data at a small-scale level in Belgium, which permitted the identification of several areas with exceptionally high crash data.…”

Gis Tools for Analyzing Accidents and Road Design: A Review

“…The attribute similarity severity index of two points is defined as the difference between each value and the value of the global average. Pirdavani et al (2014) developed crash prediction models using geographically weighted regression. It was carried out by computing Moran's I for dependent and selected explanatory variables.…”