What does this research mean for the field?

A two-stage variable selection method integrating the random forest algorithm and data augmentation improves the robustness of variable importance calculation in data envelopment analysis when the number of decision-making units is small. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

This research aims to enhance variable selection in data envelopment analysis (DEA) through a two-stage method using random forest.

March 29, 2026Decision Science Letters1 citationsOpen Access

Variable selection in data envelopment analysis: A random forest approach with data augmentation

TCTzu-Pu Chang

Key Points

This research aims to enhance variable selection in data envelopment analysis (DEA) through a two-stage method using random forest.
Implemented a baseline DEA model to classify decision making units (DMUs) as efficient or inefficient.
Trained a random forest model for high-dimensional variable selection to classify efficient and inefficient DMUs.
Calculated permutation importance indices from random forest to assess variable significance.
Addressed data augmentation to improve the robustness of permutation importance when DMUs are few.
Achieved effective classification of DMUs as either efficient or inefficient using the proposed method.
Improved the robustness of permutation importance calculations through data augmentation techniques.

Abstract

Variable selection is an important step in data envelopment analysis (DEA) when the number of decision making units (DMUs) is insufficient. This research thus proposes a two-stage variable selection method integrating a well-known supervised machine learning technique, the random forest algorithm. In the first stage, a baseline DEA model with full input and output variables is implemented and each DMU can be determined as being efficient or inefficient. In the second stage, a random forest is trained to learn how to classify efficient or inefficient DMUs well in high dimensions. Accordingly, the importance of each variable can be calculated based on permutation importance indices in random forest. This paper further discusses two issues about data augmentation in order to improve the robustness of permutation importance when the number of DMUs is quite small.

Ask AI

Helpful

Bookmark

View Full Paper