calibrated_geographically_weighted_ridge_regression
https://github.com/bishoykelleny/calibrated_geographically_weighted_ridge_regression
Science Score: 57.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
✓CITATION.cff file
Found CITATION.cff file -
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
✓DOI references
Found 3 DOI reference(s) in README -
○Academic publication links
-
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (4.9%) to scientific vocabulary
Repository
Basic Info
- Host: GitHub
- Owner: BishoyKelleny
- License: agpl-3.0
- Default Branch: main
- Size: 42 KB
Statistics
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
- Releases: 0
Metadata Files
README.md
Calibrated Geographically Weighted Ridge Regression
This program is craeted on top of the on top of mgwr: A Python implementation of multiscale geographically weighted regression for investigating process spatial heterogeneity and scale [1], while adjusting the way the spatial interaction between the OD-flow entities is accounted for in the model. In mgwr, the entities accounted for in the model are either points or polygons, with single pair of coordinates, thus, the d{ij} is typically the Euclidean distance. In our OD-flow implementation, we adapt the pair-wise spatial weighting protocol in the GWR model according to the spatial flow interaction introduced in Kordi et al. (2016) and “associated with two geographic locations, an origin and a destination” [2]. Thus, in our pair-wise spatial weighting of AAWD OD-flow, each entity encompasses a four-dimensional vector of coordinates (xi,\ yi,\ {x\prime}i,\ {y\prime}_i) as shown in Figure 65. The pair-wise weight is then formulized. The final hyperparameter in the model is the bandwidth, which is crucial for the KNN implementation of the model. As we have no prior knowledge on the extent of the spatial heterogenous behavior in the system, the model’s bandwidth is calibrated using the default mgwr “golden search optimization routine and a corrected Akaike information criterion AICc”. This optimization routine guarantees the model’s parsimony as it searches for the bandwidth corresponding to minimum AICc, consequently “penalizes smaller bandwidths”, as can be understood from the AICc formula.
[1] Oshan, T.M.; Li, Z.; Kang, W.; Wolf, L.J.; Fotheringham, A.S. mgwr: A Python Implementation of Multiscale Geographically Weighted Regression for Investigating Process Spatial Heterogeneity and Scale. ISPRS Int. J. Geo-Inf. 2019, 8, 269. https://doi.org/10.3390/ijgi8060269
[2] Maryam Kordi & A. Stewart Fotheringham (2016) Spatially Weighted Interaction Models (SWIM), Annals of the American Association of Geographers, 106:5, 990-1012, DOI: 10.1080/24694452.2016.1191990
Citation (CITATION.cff)
cff-version: 1.2.0 message: "If you use this software, please cite it as below." authors: - family-names: "Kelleny" given-names: "Bishoy" orcid: "https://orcid.org/0000-0003-1927-758X" title: "Calibrated_Geographically_Weighted_Ridge_Regression" version: 1.0.0 doi: 10.5281/zenodo.1234 date-released: 2022-02-17 url: "https://github.com/BishoyKelleny/Calibrated_Geographically_Weighted_Ridge_Regression"