8 papers
Set-Valued Control Functions
Sukjin Han, Hiroaki Kaido
The control function approach allows the researcher to identify various causal effects of interest. While powerful, it requires a strong invertibility assumption in the selection p…
Testing Information Ordering for Strategic Agents
Sukjin Han, Hiroaki Kaido, Lorenzo Magnolfi
Specifying the information structure in strategic environments is difficult for empirical researchers. We develop a test of information ordering that examines whether the true info…
How Well Do LLMs Predict Human Behavior? A Measure of their Pretrained Knowledge
Wayne Gao, Sukjin Han, Annie Liang
Large language models (LLMs) are increasingly used to predict human behavior. We propose a measure for evaluating how much knowledge a pretrained LLM brings to such a prediction: i…
Synthetic Blips: Generalizing Synthetic Controls for Dynamic Treatment Effects
Anish Agarwal, Sukjin Han, Dwaipayan Saha +2
We propose a generalization of the synthetic control and interventions methods to the setting with dynamic treatment effects. We consider the estimation of unit-specific treatment…
Copyright and Competition: Estimating Supply and Demand with Unstructured Data
Sukjin Han, Kyungho Lee
We study the competitive and welfare effects of copyright in creative industries in the face of cost-reducing technologies such as generative artificial intelligence. Creative prod…
Mining Causality: AI-Assisted Search for Instrumental Variables
Sukjin Han
The instrumental variables (IVs) method is a leading empirical strategy for causal inference. Finding IVs is a heuristic and creative process, and justifying its validity -- especi…