The Power and Limits of Predictive Approaches to Observational-Data-Driven Optimization

By Dimitris J. Bertsimas, Nathan Kallus. 2016

19 citations recorded by the source corpus.

While data-driven decision-making is transforming modern operations, most large-scale data is of an observational nature, such as transactional records. These data pose unique challenges in a variety of operational problems posed as stochastic optimization problems, including pricing and inventory management, where one must evaluate the effect of a decision, such as price or order quantity, on an uncertain cost/reward variable, such as demand, based on historical data where decision and outcome may be confounded. Often, the data lacks the features necessary to enable sound assessment of causal effects and/or the strong assumptions necessary may be dubious. Nonetheless, common practice is to assign a decision an objective value equal to the best prediction of cost/reward given the observation of the decision in the data. While in general settings this identification is spurious, for optimization purposes it is only the objective value of the final decision that matters, rather than the validity of any model used to arrive at it. In this paper, we formalize this statement in the case of observational-data-driven optimization and study both the power and limits of predictive approaches to observational-data-driven optimization with a particular focus on pricing. We provide rigorous bounds on optimality gaps of such approaches even when optimal decisions cannot be identified from data. To study potential limits of predictive approaches in real datasets, we develop a new hypothesis test for causal-effect objective optimality. Applying it to interest-rate-setting data, we empirically demonstrate that predictive approaches can be powerful in practice but with some critical limitations.

Subfields: Management Science and Operations Research, Statistics and Probability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Management Science and Operations Research

In Statistics and Probability

Citations in the corpus

Balanced Policy Evaluation and Learning (2017, background): “The aim is to transform electronic health records to personalized treatment regimes [6], transactional records to personalized pricing strategies [5], and click- and “like”-streams to personalized advertising campaigns [8] – problems of great practical significance.”
Optimization of Tree Ensembles (2017, methodology): “This issue has been the focus of some recent work in prescriptive analytics (see Bertsimas and Kallus 2016, Kallus 2016).”
From Predictive to Prescriptive Analytics (2014, background): “However, there may be many joint distributions of (X, {Y(z) : z ∈ ]}), each giving rise to a different solution z∗(x) in problem (19), that all agree with a given single joint distribution of (X,Y # Y(Z),Z) under some choice of distribution for Z (cf. Bertsimas and Kallus 2016).”