Analysis of Land Development Drivers Using Geographically Weighted Ridge Regression
Land development processes are driven by complex interactions between socio-economic and spatial factors. Acquiring an understanding of such processes and the underlying procedures helps urban and regional planners, environmental scientists, and policy makers to base their decisions on valid and pro...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-03-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/13/7/1307 |
| _version_ | 1827696434775851008 |
|---|---|
| author | Pariya Pourmohammadi Michael P. Strager Michael J. Dougherty Donald A. Adjeroh |
| author_facet | Pariya Pourmohammadi Michael P. Strager Michael J. Dougherty Donald A. Adjeroh |
| author_sort | Pariya Pourmohammadi |
| collection | DOAJ |
| description | Land development processes are driven by complex interactions between socio-economic and spatial factors. Acquiring an understanding of such processes and the underlying procedures helps urban and regional planners, environmental scientists, and policy makers to base their decisions on valid and profound information. In this work, remote-sensing-derived land-cover data were used to characterize the patterns of land development from the beginning of 1985 to the beginning of 2015, in the state of West Virginia (WV), US. We applied spatial pattern analysis, ridge regression, and Geographically Weighted Ridge Regression (GWRR) to examine the impact of population, energy resources, existing land developments dynamics, and economic status on land transformation. We showed that in presence of multicollinearity of explanatory variables, how penalizing regression models in both local and global levels lead to a better fit and decreases the model’s variance. We used geographical error analysis of regression models to visualize the difference between the model estimates and actual values. The findings of this research indicate that because of shifting geography of opportunities, the patterns and processes of land development in the studied region are unstable. This leads to fragmented land developments and prevents formation of large communities. |
| first_indexed | 2024-03-10T12:47:34Z |
| format | Article |
| id | doaj.art-906d6f12b98645b3bac2f5aa60e8d8b7 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T12:47:34Z |
| publishDate | 2021-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-906d6f12b98645b3bac2f5aa60e8d8b72023-11-21T13:21:54ZengMDPI AGRemote Sensing2072-42922021-03-01137130710.3390/rs13071307Analysis of Land Development Drivers Using Geographically Weighted Ridge RegressionPariya Pourmohammadi0Michael P. Strager1Michael J. Dougherty2Donald A. Adjeroh3Lane Department of Computer Science and Electrical Engineering, Benjamin M. Statler College of Engineering and Mineral Resources, West Virginia University, Morgantown, WV 26506-6109, USASchool of Natural Resources, Davis College of Agriculture, Natural Resources and Design, West Virginia University, Morgantown, WV 26506-6108, USASchool of Design and Community Development, Davis College of Agriculture, Natural Resources and Design, West Virginia University, Morgantown, WV 26506-6108, USALane Department of Computer Science and Electrical Engineering, Benjamin M. Statler College of Engineering and Mineral Resources, West Virginia University, Morgantown, WV 26506-6109, USALand development processes are driven by complex interactions between socio-economic and spatial factors. Acquiring an understanding of such processes and the underlying procedures helps urban and regional planners, environmental scientists, and policy makers to base their decisions on valid and profound information. In this work, remote-sensing-derived land-cover data were used to characterize the patterns of land development from the beginning of 1985 to the beginning of 2015, in the state of West Virginia (WV), US. We applied spatial pattern analysis, ridge regression, and Geographically Weighted Ridge Regression (GWRR) to examine the impact of population, energy resources, existing land developments dynamics, and economic status on land transformation. We showed that in presence of multicollinearity of explanatory variables, how penalizing regression models in both local and global levels lead to a better fit and decreases the model’s variance. We used geographical error analysis of regression models to visualize the difference between the model estimates and actual values. The findings of this research indicate that because of shifting geography of opportunities, the patterns and processes of land development in the studied region are unstable. This leads to fragmented land developments and prevents formation of large communities.https://www.mdpi.com/2072-4292/13/7/1307land development variablesmulticollinearity analysisGeographically Weighted Ridge Regression (GWRR) |
| spellingShingle | Pariya Pourmohammadi Michael P. Strager Michael J. Dougherty Donald A. Adjeroh Analysis of Land Development Drivers Using Geographically Weighted Ridge Regression Remote Sensing land development variables multicollinearity analysis Geographically Weighted Ridge Regression (GWRR) |
| title | Analysis of Land Development Drivers Using Geographically Weighted Ridge Regression |
| title_full | Analysis of Land Development Drivers Using Geographically Weighted Ridge Regression |
| title_fullStr | Analysis of Land Development Drivers Using Geographically Weighted Ridge Regression |
| title_full_unstemmed | Analysis of Land Development Drivers Using Geographically Weighted Ridge Regression |
| title_short | Analysis of Land Development Drivers Using Geographically Weighted Ridge Regression |
| title_sort | analysis of land development drivers using geographically weighted ridge regression |
| topic | land development variables multicollinearity analysis Geographically Weighted Ridge Regression (GWRR) |
| url | https://www.mdpi.com/2072-4292/13/7/1307 |
| work_keys_str_mv | AT pariyapourmohammadi analysisoflanddevelopmentdriversusinggeographicallyweightedridgeregression AT michaelpstrager analysisoflanddevelopmentdriversusinggeographicallyweightedridgeregression AT michaeljdougherty analysisoflanddevelopmentdriversusinggeographicallyweightedridgeregression AT donaldaadjeroh analysisoflanddevelopmentdriversusinggeographicallyweightedridgeregression |