Optimal Prescriptive Trees
Name
a0e810db04775b5c5a627724243dd62d608b.pdf
Description
Accepted version
Size
884.82 KB
Format
Adobe PDF
Checksum (MD5)
8a3f43c4ccdae1f709b3f367daa0287f
Author(s)
Bertsimas, Dimitris J
Dunn, Jack William
Mundru, Nishanth.
Date Issued
April 2019
Journal
INFORMS Journal on Optimization
Publisher
Institute for Operations Research and the Management Sciences (INFORMS)
Citation
Bertsimas, Dimitris et al. "Optimal Prescriptive Trees." INFORMS Journal of Optimization 1, 2 (April 2019): 164-183.
Version
Author's final manuscript
Abstract
Motivated by personalized decision making, given observational data {(xᵢ,yᵢ,zᵢ)}ⁿᵢ=1 involving features xᵢ∊ℝᵈ, assigned treatments or prescriptions zᵢ∊{1,...,𝑚}, and outcomes yᵢ∊ℝ, we propose a tree-based algorithm called optimal prescriptive tree (OPT) that uses either constant or linear models in the leaves of the tree to predict the counterfactuals and assign optimal treatments to new samples. We propose an objective function that balances optimality and accuracy. OPTs are interpretable and highly scalable, accommodate multiple treatments, and provide high-quality prescriptions. We report results involving synthetic and real data that show that OPTs either outperform or are comparable with several state-of-the-art methods. Given their combination of interpretability, scalability, generalizability, and performance, OPTs are an attractive alternative for personalized decision making in a variety of areas, such as online advertising and personalized medicine.
MIT Department
Sloan School of Management
Massachusetts Institute of Technology. Operations Research Center
Terms of Use
Creative Commons Attribution-Noncommercial-Share Alike
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1287/IJOO.2018.0005
