no code implementations • 28 Feb 2023 • Max Biggs, Georgia Perakis
At an extreme, we prove that this results in ideal formulations for tree ensembles modeling a one-dimensional feature vector.
no code implementations • 16 Feb 2022 • Max Biggs
This is in contrast to the well-studied setting in which samples of the customer's valuation (willingness to pay) are observed.
1 code implementation • 8 Dec 2021 • Ruijiang Gao, Max Biggs, Wei Sun, Ligong Han
We approach this task as a domain adaptation problem and propose a self-training algorithm which imputes outcomes with categorical values for finite unseen actions in the observational data to simulate a randomized trial through pseudolabeling, which we refer to as Counterfactual Self-Training (CST).
no code implementations • 18 Nov 2021 • Max Biggs, Ruijiang Gao, Wei Sun
The goal of this paper is to formulate loss functions that can be used for evaluating pricing policies directly from observational data, rather than going through an intermediate demand estimation stage, which may suffer from bias.
no code implementations • 1 Jan 2021 • Ruijiang Gao, Max Biggs, Wei Sun, Ligong Han
We approach this task as a domain adaptation problem and propose a self-training algorithm which imputes outcomes for the unseen actions in the observational data to simulate a randomized trial.
no code implementations • 3 Jul 2020 • Max Biggs, Wei Sun, Markus Ettl
Data-driven pricing strategies are becoming increasingly common, where customers are offered a personalized price based on features that are predictive of their valuation of a product.