Forest Guided Smoothing

TL;DR

This paper introduces a forest-guided smoothing technique that leverages random forest outputs to create interpretable, spatially adaptive local smoothers suitable for bias correction, confidence intervals, and structural analysis, demonstrated on synthetic and Covid-19 data.

Contribution

It presents a novel, simple linear smoothing method derived from random forests, enabling interpretability and application in bias correction and variable importance assessment.

Findings

Effective on synthetic examples

Applied successfully to Covid-19 data

Enables bias correction and confidence interval estimation

Abstract

We use the output of a random forest to define a family of local smoothers with spatially adaptive bandwidth matrices. The smoother inherits the flexibility of the original forest but, since it is a simple, linear smoother, it is very interpretable and it can be used for tasks that would be intractable for the original forest. This includes bias correction, confidence intervals, assessing variable importance and methods for exploring the structure of the forest. We illustrate the method on some synthetic examples and on data related to Covid-19.