eks-package {eks}

eks

Description

This package extends the functionality of the kernel smoothing functions from the ks package in base R to the tidyverse and to GIS (Geographical Information Systems) ecosystems.

Details

As the kernel smoothers from the ks package are prefixed as k*, their equivalents in eks are systematically named as follows:

• tidy_k* for 1- and 2-d tidy data

• st_k* for 2-d geospatial data.

The output data tibbles (tidy data frames provided by the tibble package) from tidy_k* can be visualised within the ggplot2 graphical interface, using the usual layer functions and the custom ones supplied in this package. These tidy_k* functions are analogous to those in the broom and related packages, though the latter tend to focus on tidying the summary diagnostic output from model fitting (and not on tidying the underlying estimates themselves), whereas tidy_k* are more substantive since they do compute tidy estimates.

The output simple feature geometries (provided by the sf package) from st_k* can be visualised in the (i) ggplot2 graphical interface using primarily the geom_sf layer function, or (ii) in the base R graphical interface using the plot method supplied in this package. These simple feature geometries can also be exported as standard geospatial formats (e.g. shapefile, GEOS geometry) for use in external GIS software such as ArcGIS and QGIS.

Author(s)

Tarn Duong

References

Chacon, J.E. & Duong, T. (2018) Multivariate Kernel Smoothing and Its Applications. Chapman & Hall/CRC, Boca Raton.

Duong, T. (2022) Statistical visualisation for tidy and geospatial data in R via kernel smoothing methods in the eks package. Submitted. DOI:10.48550/arXiv.2203.01686