README.md

In XiaosuTong/Spaloess: Spatially Local Polynomial Regression

Spatial Locally Weighted Regression

This package contains enhancements to the loess implementation that comes with base R. Most of computation in FORTRAN are kept as same in loess function except the distance calculation. Modification is made in spaloess.f file.

Installation

# from github
devtools::install_github("XiaosuTong/Spaloess")

Adding Feature

Here are some of the added features over loess: - Can predict NA directly without calling predict.loess after loess to get fitted value at NA locations - Add distance argument which can be different type of distance function: "Euclid", "Latlong" for great circle distance - Add a function which can generate kd-tree from a dataset. - Includes all locations to build the kd-tree. Argument alltree controls this.

XiaosuTong/Spaloess documentation built on May 9, 2019, 11:06 p.m.