Are ROC-improved logistic regressions beneficial to statistical models of land change? A counter example

Logistic regression (LR) is widely used in land change modelling; however, its traditional form assumes independent input variables, which is often not realistic. Although the improved models offer better fitting capabilities, it is unclear whether this leads to more accurate land change simulations...

Full description

Saved in:
Bibliographic Details
Main Authors: Mingyang Qin, Yiru Xie, Shi Hua, Zhanhong Liu, Peichao Gao
Format: Article
Language:English
Published: Taylor & Francis Group 2025-07-01
Series:Annals of GIS
Subjects:
Online Access:https://www.tandfonline.com/doi/10.1080/19475683.2025.2523736
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849421985393672192
author Mingyang Qin
Yiru Xie
Shi Hua
Zhanhong Liu
Peichao Gao
author_facet Mingyang Qin
Yiru Xie
Shi Hua
Zhanhong Liu
Peichao Gao
author_sort Mingyang Qin
collection DOAJ
description Logistic regression (LR) is widely used in land change modelling; however, its traditional form assumes independent input variables, which is often not realistic. Although the improved models offer better fitting capabilities, it is unclear whether this leads to more accurate land change simulations. To address this gap, we compared the basic LR model with five classic improved models using Lhasa as a case study, comparing the receiver-operating characteristic (ROC) values of each model and further evaluating the performance of the land change models generated by coupling each LR model with the CLUE-s model using six evaluation metrics (Kno, Klocation, Kquantity and the divergence indices (D, A and Q)). The results show that the improved LR models exhibit significantly enhanced ROC values. Specifically, the combined LR achieved the highest average ROC value of 0.941 across different neighbourhood sizes, and the average ROC values of all improved regressions exceeded 0.9, which is significantly higher than that of the ordinary LR (0.872), which remains unaffected by changes in neighbourhood size. However, concerning the land change simulation accuracy, ordinary LR demonstrated a clear advantage, consistently achieving the best performance across all six evaluation metrics regardless of neighbourhood size. Conversely, the improved regressions performed worse, and the combined logistic regression (CL), despite having the highest ROC, performed the poorest in four out of the six evaluation metrics. This finding indicates that there is no inherent link between higher ROC values and improved land change model accuracy. This study further explores the underlying causes of this phenomenon and suggests directions for improvement.
format Article
id doaj-art-0afe155f08ae44af984e8e92823cb207
institution Kabale University
issn 1947-5683
1947-5691
language English
publishDate 2025-07-01
publisher Taylor & Francis Group
record_format Article
series Annals of GIS
spelling doaj-art-0afe155f08ae44af984e8e92823cb2072025-08-20T03:31:19ZengTaylor & Francis GroupAnnals of GIS1947-56831947-56912025-07-0112110.1080/19475683.2025.2523736Are ROC-improved logistic regressions beneficial to statistical models of land change? A counter exampleMingyang Qin0Yiru Xie1Shi Hua2Zhanhong Liu3Peichao Gao4State Key Laboratory of Earth Surface Processes and Disaster Risk Reduction, Faculty of Geographical Science, Beijing Normal University, Beijing, ChinaBeijing Institute of Surveying and Mapping, Beijing, ChinaState Key Laboratory of Earth Surface Processes and Disaster Risk Reduction, Faculty of Geographical Science, Beijing Normal University, Beijing, ChinaState Key Laboratory of Earth Surface Processes and Disaster Risk Reduction, Faculty of Geographical Science, Beijing Normal University, Beijing, ChinaState Key Laboratory of Earth Surface Processes and Disaster Risk Reduction, Faculty of Geographical Science, Beijing Normal University, Beijing, ChinaLogistic regression (LR) is widely used in land change modelling; however, its traditional form assumes independent input variables, which is often not realistic. Although the improved models offer better fitting capabilities, it is unclear whether this leads to more accurate land change simulations. To address this gap, we compared the basic LR model with five classic improved models using Lhasa as a case study, comparing the receiver-operating characteristic (ROC) values of each model and further evaluating the performance of the land change models generated by coupling each LR model with the CLUE-s model using six evaluation metrics (Kno, Klocation, Kquantity and the divergence indices (D, A and Q)). The results show that the improved LR models exhibit significantly enhanced ROC values. Specifically, the combined LR achieved the highest average ROC value of 0.941 across different neighbourhood sizes, and the average ROC values of all improved regressions exceeded 0.9, which is significantly higher than that of the ordinary LR (0.872), which remains unaffected by changes in neighbourhood size. However, concerning the land change simulation accuracy, ordinary LR demonstrated a clear advantage, consistently achieving the best performance across all six evaluation metrics regardless of neighbourhood size. Conversely, the improved regressions performed worse, and the combined logistic regression (CL), despite having the highest ROC, performed the poorest in four out of the six evaluation metrics. This finding indicates that there is no inherent link between higher ROC values and improved land change model accuracy. This study further explores the underlying causes of this phenomenon and suggests directions for improvement.https://www.tandfonline.com/doi/10.1080/19475683.2025.2523736CLUE-Slogistic regressionTibetland-use change simulation
spellingShingle Mingyang Qin
Yiru Xie
Shi Hua
Zhanhong Liu
Peichao Gao
Are ROC-improved logistic regressions beneficial to statistical models of land change? A counter example
Annals of GIS
CLUE-S
logistic regression
Tibet
land-use change simulation
title Are ROC-improved logistic regressions beneficial to statistical models of land change? A counter example
title_full Are ROC-improved logistic regressions beneficial to statistical models of land change? A counter example
title_fullStr Are ROC-improved logistic regressions beneficial to statistical models of land change? A counter example
title_full_unstemmed Are ROC-improved logistic regressions beneficial to statistical models of land change? A counter example
title_short Are ROC-improved logistic regressions beneficial to statistical models of land change? A counter example
title_sort are roc improved logistic regressions beneficial to statistical models of land change a counter example
topic CLUE-S
logistic regression
Tibet
land-use change simulation
url https://www.tandfonline.com/doi/10.1080/19475683.2025.2523736
work_keys_str_mv AT mingyangqin arerocimprovedlogisticregressionsbeneficialtostatisticalmodelsoflandchangeacounterexample
AT yiruxie arerocimprovedlogisticregressionsbeneficialtostatisticalmodelsoflandchangeacounterexample
AT shihua arerocimprovedlogisticregressionsbeneficialtostatisticalmodelsoflandchangeacounterexample
AT zhanhongliu arerocimprovedlogisticregressionsbeneficialtostatisticalmodelsoflandchangeacounterexample
AT peichaogao arerocimprovedlogisticregressionsbeneficialtostatisticalmodelsoflandchangeacounterexample