Regression with Missing Data, a Comparison Study of TechniquesBased on Random Forests
This work addresses the challenge of missing data in statistical modeling for researchers and practitioners, but it is incremental as it builds on prior consistency results and focuses on comparative performance.
The paper tackles the problem of handling missing data in regression using random forests by comparing various techniques, including a new algorithm, and finds that it performs competitively with existing methods in terms of quadratic errors and bias across different missing value mechanisms.
In this paper we present the practical benefits of a new random forest algorithm to deal withmissing values in the sample. The purpose of this work is to compare the different solutionsto deal with missing values with random forests and describe our new algorithm performanceas well as its algorithmic complexity. A variety of missing value mechanisms (such as MCAR,MAR, MNAR) are considered and simulated. We study the quadratic errors and the bias ofour algorithm and compare it to the most popular missing values random forests algorithms inthe literature. In particular, we compare those techniques for both a regression and predictionpurpose. This work follows a first paper Gomez-Mendez and Joly (2020) on the consistency ofthis new algorithm.