Bootstrap robust prescriptive analytics

DOI10.1007/S10107-021-01679-2zbMATH Open1504.90076arXiv1711.09974OpenAlexW3177107840MaRDI QIDQ2089765FDOQ2089765

Authors: Dimitris Bertsimas, Bart P. G. Van Parys 

Publication date: 24 October 2022

Published in: Mathematical Programming. Series A. Series B (Search for Journal in Brave)

Abstract: We address the problem of prescribing an optimal decision in a framework where the cost function depends on uncertain problem parameters that need to be learned from data. Earlier work proposed prescriptive formulations based on supervised machine learning methods. These prescriptive methods can factor in contextual information on a potentially large number of covariates to take context specific actions which are superior to any static decision. When working with noisy or corrupt data, however, such nominal prescriptive methods can be prone to adverse overfitting phenomena and fail to generalize on out-of-sample data. In this paper we combine ideas from robust optimization and the statistical bootstrap to propose novel prescriptive methods which safeguard against overfitting. We show indeed that a particular entropic robust counterpart to such nominal formulations guarantees good performance on synthetic bootstrap data. As bootstrap data is often a sensible proxy to actual out-of-sample data, our robust counterpart can be interpreted to directly encourage good out-of-sample performance. The associated robust prescriptive methods furthermore reduce to convenient tractable convex optimization problems in the context of local learning methods such as nearest neighbors and Nadaraya-Watson learning. We illustrate our data-driven decision-making framework and our novel robustness notion on a small newsvendor problem.

Full work available at URL: https://arxiv.org/abs/1711.09974

Recommendations

• Robust sample average approximation
• Data-driven robust optimization
• Calibration of distributionally robust empirical optimization models
• Likelihood robust optimization for data-driven problems
• Newsvendor problems: an integrated method for estimation and optimisation

zbMATH Keywords

data analyticsdistributionally robust optimizationstatistical bootstrapNadaraya-Watson learningnearest neighbors learning

Mathematics Subject Classification ID

Convex programming (90C25) Stochastic programming (90C15) Robustness in mathematical programming (90C17)

Cites Work

• Julia: a fresh approach to numerical computing
• The elements of statistical learning. Data mining, inference, and prediction
• Random forests
• Title not available (Why is that?)
• Title not available (Why is that?)
• On Information and Sufficiency
• Title not available (Why is that?)
• A distribution-free theory of nonparametric regression
• Title not available (Why is that?)
• Distributionally robust joint chance constraints with second-order moment information
• Ambiguous chance constrained problems and robust optimization
• Generalized Gauss inequalities via semidefinite programming
• Distributionally robust optimization under moment uncertainty with application to data-driven problems
• A Semidefinite Programming Approach to Optimal-Moment Bounds for Convex Classes of Distributions
• Sanov property, generalized I-projection and a conditional limit theorem
• Likelihood robust optimization for data-driven problems
• Multivariate locally adaptive density estimation.
• Ambiguity in portfolio selection
• Distributionally Robust Convex Optimization
• Data-driven distributionally robust optimization using the Wasserstein metric: performance guarantees and tractable reformulations
• Convergence analysis for distributionally robust optimization and equilibrium problems
• Computationally tractable counterparts of distributionally robust constraints on risk measures
• The big data newsvendor: practical insights from machine learning
• Strong laws of large numbers and nonparametric estimation

Cited In (6)

• Data-Driven Approximation of Contextual Chance-Constrained Stochastic Programs
• Rollout-based routing strategies with embedded prediction: a fish trawling application
• Residuals-based distributionally robust optimization with covariate information
• Coupled learning enabled stochastic programming with endogenous uncertainty
• Frameworks and results in distributionally robust optimization
• Contextual robust optimisation with uncertainty quantification

Uses Software

• ECOS
• Julia
• Convex.jl

This page was built for publication: Bootstrap robust prescriptive analytics

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2089765)