Summary
Contents
Subject index
This is a proposal for a core text in geocomputation; the use of computational techniques to model and solve problems – population, environment, planning, etc etc - in spatial context. It's multi-disciplinary and inter-disciplinary so this text is a generic primer that looks at: visualisation and exploratory spatial data analysis; space time modeling; spatial algorithms; spatial regression and statistics; and decision making. The edited text is organized in five sections: 1. Introducing Applied Geocomputation; 2. Describing how the world looks; 3. Exploring movements in space; 4. Making geographical decisions; and 5. Explaining how the world works
Geographically Weighted Generalised Linear Modelling
Geographically Weighted Generalised Linear Modelling
Introduction
One of the most popular approaches to statistical model building is regression analysis. Here, we would associate a response variable we wish to predict with explanatory variables, using an assumed relationship such as a linear function. We fit the models to the observed data statistically, and then use the model to infer the properties of underlying processes hidden within the dataset. This chapter focuses on a specific type of regression modelling for spatial analyses, namely geographically weighted regression (GWR), and its extended form, semi-parametric geographically weighted generalised linear modelling (S-GWGLM). Such models replicate the geographically varying aspects of the association between the response and explanatory variables. The S-GWGLM modelling framework is implemented in ...
- Loading...