A diffusion-based kernel density estimator (diffKDE, version 1) with optimal bandwidth approximation for the analysis of data in geoscience and ecological research

Pelz, Maria-Theresia; Schartau, Markus; Somes, Christopher J.; Lampe, Vanessa; Slawig, Thomas

doi:https://doi.org/10.5194/gmd-16-6609-2023

Articles | Volume 16, issue 22

https://doi.org/10.5194/gmd-16-6609-2023

© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/gmd-16-6609-2023

© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 16, issue 22

Methods for assessment of models

|

16 Nov 2023

Methods for assessment of models |

| 16 Nov 2023

A diffusion-based kernel density estimator (diffKDE, version 1) with optimal bandwidth approximation for the analysis of data in geoscience and ecological research

Maria-Theresia Pelz, Markus Schartau, Christopher J. Somes, Vanessa Lampe, and Thomas Slawig

Data sets

A global marine particulate organic carbon-13 isotope data product Maria-Theresia Verwega, Christopher J. Somes, Robyn E. Tuerena, Anne Lorrain https://doi.org/10.1594/PANGAEA.929931

Model code and software

Diffusion-based kernel density estimator (diffKDE) M.-T. Pelz and T. Slawig https://doi.org/10.5281/ZENODO.7594915

Short summary

Kernel density estimators (KDE) approximate the probability density of a data set without the assumption of an underlying distribution. We used the solution of the diffusion equation, and a new approximation of the optimal smoothing parameter build on two pilot estimation steps, to construct such a KDE best suited for typical characteristics of geoscientific data. The resulting KDE is insensitive to noise and well resolves multimodal data structures as well as boundary-close data.