Residuals-Based Contextual Distributionally Robust Optimization with Decision-Dependent Uncertainty: Theoretical Guarantees and Decomposition Algorithm

TL;DR

This paper introduces a residuals-based distributionally robust optimization model that accounts for decision-dependent uncertainty, providing theoretical guarantees and a specialized decomposition algorithm for efficient solution.

Contribution

It develops a novel DRO framework with decision-dependent ambiguity sets using residuals, and proposes a finite convergence Bender's decomposition algorithm.

Findings

The model achieves asymptotic optimality and finite sample guarantees.

The specialized algorithm converges in finite steps.

Numerical experiments demonstrate the approach's effectiveness.

Abstract

We consider a residuals-based distributionally robust optimization (DRO) model, where the underlying uncertainty depends on both covariate information and our decisions. We adopt both parametric and nonparametric regression models to learn the latent decision dependency and construct a nominal distribution (thereby ambiguity sets) around the learned model using empirical residuals from the regressions. We formulate the ambiguity set via the Wasserstein distance, where the nominal distribution is both decision- and covariate-dependent. We provide conditions under which desired statistical properties such as asymptotic optimality, rate of convergence, and finite sample guarantees are satisfied. To solve the resulting DRO model, we develop a specialized Bender's decomposition algorithm with nonlinear cuts and prove its finite convergence. Through numerical experiments, we illustrate the effectiveness of our approach and the benefits of integrating decision dependency into a residuals-based DRO framework.