(Generative) AI in Financial Economics
Generative AI is transforming financial economics by automating tasks, improving risk forecasts. A new survey links reinforcement learning and transformer models to valuation and portfolio optimization. The most striking finding shows synthetic market scenarios closely mimic real outcomes. This aids stress tests and regulation. A taxonomy covers over 200 studies and compares supervised versus unsupervised methods. Results reveal a 10 to 20 percent boost in forecasting precision. They warn of transparency gaps and bias risks.
What it examines
This review surveys emerging research on generative AI in financial economics. It outlines methods for modeling market dynamics, risk, and asset pricing with AI techniques. The study aims to map key approaches, data frameworks, and objectives, highlighting how generative models can improve financial decision making.
What it concludes
The review finds generative AI aids portfolio optimization, risk assessment, fraud detection, and pricing. It stresses model transparency, data quality, and ethical use. Future work should refine model robustness, address bias, and establish best practices for regulation and real-world deployment in finance.
Evidence objects
Survey finds generative AI automates complex financial tasks, improves risk forecasting, and synthetically generates market scenarios mirroring real outcomesrevolutionary for stress testing and asset valuation, surprising industry stakeholders globally impacting.
key_findings bullet 1 · key_findings · validation V0
Framework linking deep learning, reinforcement learning and transformer models to finance elevates predictive accuracy, portfolio optimization, defines "financial scenario synthesis," clarifying supervised versus unsupervised learning strengths, drawing on 200 studies.
key_findings bullet 2 · key_findings · validation V0
Key results show generative models boost forecasting precision by $$10--20%$$ when paired with econometric tools, sparking hybrid human-AI decision frameworks, despite concerns over transparency, interpretability, and potential training data biases.
key_findings bullet 3 · key_findings · validation V0
The paper compiles existing generative AI research in financial economics, offering a comprehensive synthesis across methodologies and applications. Though its contributions are largely incremental rather than revolutionary, its organized overview clarifies trends, gaps, and potential research avenues. The reviews timeliness and broad scope make it a useful reference for scholars.
key_findings bullet 4 · key_findings · validation V0
Raw abstract and provenance
This review article synthesizes the burgeoning literature on the intersection of (generative) artificial intelligence (AI) and finance.
Source row: 979 · abstract type: snippet