earthUI

An Interactive Interface for the earth (MARS) Package

Authors

DOI:

https://doi.org/10.67330/7fj41b65

Keywords:

MARS, multivariate adaptive regression splines, earth, R, graphical user interface, real estate appraisal

Abstract

earthUI is a graphical user interface for the R earth package, which implements Multivariate Adaptive Regression Splines (MARS). It offers three purpose modes—general predictive modeling, real-estate appraisal, and market-area analysis—and guides the user through data import, model configuration, fitting, and the interpretation of earth's diagnostics and graphical output. This article documents earthUI's data-format requirements, modeling workflow, output displays, and complete feature reference.

Downloads

Download data is not yet available.

Author Biography

  • William Bert Craytor

    Bert Craytor is a real estate valuation consultant specializing in statistical modeling and automated valuation methods. His work focuses on applying regression-based techniques — including MARS, GLM and GAM models — to residential and commercial appraisal. He is the developer of the earthUI, glmnetUI, and mgcvUI packages for  R.

References

Friedman, Jerome H. 1991. “Multivariate Adaptive Regression Splines.” The Annals of Statistics 19 (1): 1–141. https://doi.org/10.1214/aos/1176347963.

Milborrow, Stephen. 2024. “Earth: Multivariate Adaptive Regression Splines.” October 1. https://CRAN.R-project.org/package=earth.

Milborrow, Stephen. n.d.-a. “Notes on the Earth Package.” http://www.milbo.org/doc/earth-notes.pdf.

Milborrow, Stephen. n.d.-b. “Variance Models in Earth.” http://www.milbo.org/doc/earth-varmod.pdf.

Downloads

Published

2026-05-23 — Updated on 2026-07-13

Versions

How to Cite

earthUI: An Interactive Interface for the earth (MARS) Package. (2026). Valuation Engineer Journal, 1(1), 97-141. https://doi.org/10.67330/7fj41b65 (Original work published 2026)