Simulating Macroeconomic Expectations in Survey Experiments with LLM-based Economic Agents
Researchers have developed a new framework using Large Language Model (LLM) based agents to simulate macroeconomic expectations in survey experiments. These LLM agents are equipped with modules for personal characteristics, prior expectations, and external information. The framework successfully replicated three survey designs, generating expectation distributions similar to human data and capturing qualitative patterns in open-ended responses, offering insights into belief gaps between AI and humans. AI
IMPACT This research provides a novel method for understanding and potentially bridging the gap between AI and human economic expectations.