A multiscale geographically weighted regression for dual-relationship analysis: case of London traffic accident analysis

The dataset and codes support a study discussed in the manuscript entitled "A multiscale geographically weighted regression for dual-relationship analysis: a case study on urban traffic accident analysis". The contents of the dataset and codes are followed: ①Code Folder: This folder stores the processing code of mgwr with London traffic accident data. The MGWR (including GWR) python modules are authored by Taylor Oshan, which can be obtained from the link https://github.com/pysal/mgwr. ②Data Folder: This folder stores the unprocessed source data, processed data and the model result. There are 3 types of source data: (1) Road safety data of Great Britain, from data.gov.uk (https://data.gov.uk/dataset/cb7ae6f0-4be6-4935-9277-47e5ce24a11f/road-safety-data); (2) The LSOA of London and English Indices of Deprivation 2015, from London Datastore (https://data.london.gov.uk/dataset); (3) Road, POI, and land use data from Open Street Map (https://www.openstreetmap.org/).