Publications
Publications
Show Results For
- All HBS Web
(52)
- Faculty Publications (8)
Page 1 of 8
Results
- 2022
- Working Paper
Statistical Inference for Heterogeneous Treatment Effects Discovered by Generic Machine Learning in Randomized Experiments
By: Kosuke Imai and Michael Lingzhi Li
Researchers are increasingly turning to machine learning (ML) algorithms to investigate causal heterogeneity in randomized experiments. Despite their promise, ML algorithms may fail to accurately ascertain heterogeneous treatment effects under practical settings with... View Details
- February 2022
- Technical Note
Ethical Analysis: Fairness
By: Nien-hê Hsieh
Concerns about fairness can arise across a wide range of contexts. They include the treatment of others, how much things cost, how much workers are paid, the outcome of a decision, and how we assign benefits and burdens across individuals. What counts as fair in a... View Details
- November 2021
- Article
Panel Experiments and Dynamic Causal Effects: A Finite Population Perspective
By: Iavor Bojinov, Ashesh Rambachan and Neil Shephard
In panel experiments, we randomly assign units to different interventions, measuring their outcomes, and repeating the procedure in several periods. Using the potential outcomes framework, we define finite population dynamic causal effects that capture the relative... View Details
- 2016
- Working Paper
Through the Grapevine: Network Effects on the Design of Executive Compensation Contracts
By: Susanna Gallani
Effective design of executive compensation contracts involves choosing and weighting performance measures, as well as defining the mix between fixed and incentive-based pay components, with a view to fostering talent retention and goal congruence. The variability in... View Details
- October 2007
- Article
The Art of Designing Markets
By: Alvin E. Roth
Traditionally, markets have been viewed as simply the confluence of supply and demand. But to function properly, they must be able to attract a sufficient number of buyers and sellers, induce participants to make their preferences clear, and overcome congestion by... View Details
- December 1996 (Revised June 1998)
- Case
Midnight Networks, Inc.
By: H. Kent Bowen and Marilyn Matis
Midnight Networks, Inc., is a small computer network validation company. This case describes how the five founders built their business from operations earnings and how they established "best practices" operational processes to run their firm successfully. Operational... View Details
- Forthcoming
- Article
Outcome-Driven Dynamic Refugee Assignment with Allocation Balancing
By: Kirk Bansak and Elisabeth Paulson
This study proposes two new dynamic assignment algorithms to match refugees and asylum seekers to geographic localities within a host country. The first, currently implemented in a multi-year pilot in Switzerland, seeks to maximize the average predicted employment... View Details
- Forthcoming
- Article
Statistical Inference for Heterogeneous Treatment Effects Discovered by Generic Machine Learning in Randomized Experiments
By: Kosuke Imai and Michael Lingzhi Li
Researchers are increasingly turning to machine learning (ML) algorithms to investigate causal heterogeneity in randomized experiments. Despite their promise, ML algorithms may fail to accurately ascertain heterogeneous treatment effects under practical settings with... View Details