Revisiting Kappa to Account for Change in the Accuracy Assessment of Land-Use Change Models. van Vliet , J., Bregt, A. K., & Hagen-Zanker, A. Ecological Modelling, 222(8):1367–1375, April, 2011.
Paper doi abstract bibtex Land-use change models are typically calibrated to reproduce known historic changes. Calibration results can then be assessed by comparing two datasets: the simulated land-use map and the actual land-use map at the same time. A common method for this is the Kappa statistic, which expresses the agreement between two categorical datasets corrected for the expected agreement. This expected agreement is based on a stochastic model of random allocation given the distribution of class sizes. However, when a model starts from an initial land-use map and makes changes to it, that stochastic model does not pose a meaningful reference level. This paper introduces KSimulation, a statistic that is identical in form to the Kappa statistic but instead applies a more appropriate stochastic model of random allocation of class transitions relative to the initial map. The new method is illustrated on a simple example and then the results of the Kappa statistic and KSimulation are compared using the results of a land-use model. It is found that only KSimulation truly tests models in their capacity to explain land-use changes over time, and unlike Kappa it does not inflate results for simulations where little change takes place over time.
@article{van_vliet_revisiting_2011,
title = {Revisiting {{Kappa}} to Account for Change in the Accuracy Assessment of Land-Use Change Models},
author = {{van Vliet}, J. and Bregt, A. K. and {Hagen-Zanker}, A.},
year = {2011},
month = apr,
journal = {Ecological Modelling},
volume = {222},
number = {8},
pages = {1367--1375},
issn = {0304-3800},
doi = {10.1016/j.ecolmodel.2011.01.017},
url = {http://www.sciencedirect.com/science/article/pii/S0304380011000494},
urldate = {2014-01-10TZ},
abstract = {Land-use change models are typically calibrated to reproduce known historic changes. Calibration results can then be assessed by comparing two datasets: the simulated land-use map and the actual land-use map at the same time. A common method for this is the Kappa statistic, which expresses the agreement between two categorical datasets corrected for the expected agreement. This expected agreement is based on a stochastic model of random allocation given the distribution of class sizes. However, when a model starts from an initial land-use map and makes changes to it, that stochastic model does not pose a meaningful reference level. This paper introduces KSimulation, a statistic that is identical in form to the Kappa statistic but instead applies a more appropriate stochastic model of random allocation of class transitions relative to the initial map. The new method is illustrated on a simple example and then the results of the Kappa statistic and KSimulation are compared using the results of a land-use model. It is found that only KSimulation truly tests models in their capacity to explain land-use changes over time, and unlike Kappa it does not inflate results for simulations where little change takes place over time.},
keywords = {Accuracy assessment,Kappa statistic,Land-use change,map comparison,Model calibration}
}
Downloads: 0
{"_id":"oKuS5tjwKSPDMJ2ZA","bibbaseid":"vanvliet-bregt-hagenzanker-revisitingkappatoaccountforchangeintheaccuracyassessmentoflandusechangemodels-2011","author_short":["van Vliet , J.","Bregt, A. K.","Hagen-Zanker, A."],"bibdata":{"bibtype":"article","type":"article","title":"Revisiting Kappa to Account for Change in the Accuracy Assessment of Land-Use Change Models","author":[{"propositions":["van Vliet"],"lastnames":[],"firstnames":["J."],"suffixes":[]},{"propositions":[],"lastnames":["Bregt"],"firstnames":["A.","K."],"suffixes":[]},{"propositions":[],"lastnames":["Hagen-Zanker"],"firstnames":["A."],"suffixes":[]}],"year":"2011","month":"April","journal":"Ecological Modelling","volume":"222","number":"8","pages":"1367–1375","issn":"0304-3800","doi":"10.1016/j.ecolmodel.2011.01.017","url":"http://www.sciencedirect.com/science/article/pii/S0304380011000494","urldate":"2014-01-10TZ","abstract":"Land-use change models are typically calibrated to reproduce known historic changes. Calibration results can then be assessed by comparing two datasets: the simulated land-use map and the actual land-use map at the same time. A common method for this is the Kappa statistic, which expresses the agreement between two categorical datasets corrected for the expected agreement. This expected agreement is based on a stochastic model of random allocation given the distribution of class sizes. However, when a model starts from an initial land-use map and makes changes to it, that stochastic model does not pose a meaningful reference level. This paper introduces KSimulation, a statistic that is identical in form to the Kappa statistic but instead applies a more appropriate stochastic model of random allocation of class transitions relative to the initial map. The new method is illustrated on a simple example and then the results of the Kappa statistic and KSimulation are compared using the results of a land-use model. It is found that only KSimulation truly tests models in their capacity to explain land-use changes over time, and unlike Kappa it does not inflate results for simulations where little change takes place over time.","keywords":"Accuracy assessment,Kappa statistic,Land-use change,map comparison,Model calibration","bibtex":"@article{van_vliet_revisiting_2011,\n title = {Revisiting {{Kappa}} to Account for Change in the Accuracy Assessment of Land-Use Change Models},\n author = {{van Vliet}, J. and Bregt, A. K. and {Hagen-Zanker}, A.},\n year = {2011},\n month = apr,\n journal = {Ecological Modelling},\n volume = {222},\n number = {8},\n pages = {1367--1375},\n issn = {0304-3800},\n doi = {10.1016/j.ecolmodel.2011.01.017},\n url = {http://www.sciencedirect.com/science/article/pii/S0304380011000494},\n urldate = {2014-01-10TZ},\n abstract = {Land-use change models are typically calibrated to reproduce known historic changes. Calibration results can then be assessed by comparing two datasets: the simulated land-use map and the actual land-use map at the same time. A common method for this is the Kappa statistic, which expresses the agreement between two categorical datasets corrected for the expected agreement. This expected agreement is based on a stochastic model of random allocation given the distribution of class sizes. However, when a model starts from an initial land-use map and makes changes to it, that stochastic model does not pose a meaningful reference level. This paper introduces KSimulation, a statistic that is identical in form to the Kappa statistic but instead applies a more appropriate stochastic model of random allocation of class transitions relative to the initial map. The new method is illustrated on a simple example and then the results of the Kappa statistic and KSimulation are compared using the results of a land-use model. It is found that only KSimulation truly tests models in their capacity to explain land-use changes over time, and unlike Kappa it does not inflate results for simulations where little change takes place over time.},\n keywords = {Accuracy assessment,Kappa statistic,Land-use change,map comparison,Model calibration}\n}\n\n","author_short":["van Vliet , J.","Bregt, A. K.","Hagen-Zanker, A."],"key":"van_vliet_revisiting_2011","id":"van_vliet_revisiting_2011","bibbaseid":"vanvliet-bregt-hagenzanker-revisitingkappatoaccountforchangeintheaccuracyassessmentoflandusechangemodels-2011","role":"author","urls":{"Paper":"http://www.sciencedirect.com/science/article/pii/S0304380011000494"},"keyword":["Accuracy assessment","Kappa statistic","Land-use change","map comparison","Model calibration"],"metadata":{"authorlinks":{}}},"bibtype":"article","biburl":"https://raw.githubusercontent.com/EECi/home/main/docs/publications/EECi.bib","dataSources":["oXrzxozuzK4m5k4ju"],"keywords":["accuracy assessment","kappa statistic","land-use change","map comparison","model calibration"],"search_terms":["revisiting","kappa","account","change","accuracy","assessment","land","use","change","models","van vliet ","bregt","hagen-zanker"],"title":"Revisiting Kappa to Account for Change in the Accuracy Assessment of Land-Use Change Models","year":2011}