Jimmy Wu

Assistant Professor of Finance

Tippie College of Business, University of Iowa

jimmy-wu@uiowa.edu Curriculum Vitae Google Scholar

About

I am an Assistant Professor of Finance at the University of Iowa. My research interests include empirical asset pricing, behavioral finance, large language models, and applications of machine learning and AI in finance.

Research

Publications

Chronologically Consistent Large Language Models

with Songrun He, Linying Lv, and Asaf Manela

Journal of Financial Economics, 2026, 185: 104374
Abstract

Large language models are increasingly used in social sciences, but their training data can introduce lookahead bias and training leakage. A good chronologically consistent language model requires efficient use of training data to maintain accuracy despite time-restricted data. Here, we overcome this challenge by training a suite of chronologically consistent large language models, ChronoBERT and ChronoGPT, which incorporate only the text data that would have been available at each point in time. Despite this strict temporal constraint, our models achieve strong performance on natural language processing benchmarks, outperforming or matching widely used models (e.g., BERT), and remain competitive with larger open-weight models. Lookahead bias is model and application-specific because even if a chronologically consistent language model has poorer language comprehension, a regression or prediction model applied on top of the language model can compensate. In an asset pricing application predicting next-day stock returns from financial news embeddings, we find that ChronoBERT and ChronoGPT’s real-time outputs achieve Sharpe ratios comparable to a much larger and potentially biased Llama model. Our results demonstrate a scalable, practical framework to mitigate training leakage, ensuring more credible backtests and predictions across finance and other social science domains.
He, Songrun, Linying Lv, Asaf Manela, and Jimmy Wu. 2026. “Chronologically Consistent Large Language Models.” Journal of Financial Economics 185: 104374. https://doi.org/10.1016/j.jfineco.2026.104374.
```
@article{He_Lv_Manela_Wu_2026_JFE,
  author  = {He, Songrun and Lv, Linying and Manela, Asaf and Wu, Jimmy},
  title   = {Chronologically Consistent Large Language Models},
  journal = {Journal of Financial Economics},
  year    = {2026},
  volume  = {185},
  pages   = {104374},
  doi     = {10.1016/j.jfineco.2026.104374}
}
```
The True Colors of Money: Racial Representation and Asset Management

with Lina Han, Xing Huang, and Ohad Kadan

Review of Financial Studies, 2026
Abstract

We examine the role of race and ethnicity in the mutual fund context at two distinct levels. At the fund manager level, we document a co-racial tilt—funds managed by minority-dominant (White-dominant) teams allocate larger portfolio weights to minority-led (White-led) firms. This tilt is not associated with superior performance. It diminishes as fund managers gain experience, suggesting the presence of inaccurate statistical discrimination. At the investor level, we find that minority-dominant funds are penalized similarly to White-dominant funds for poor performance but are not rewarded as much for superior performance. Overall, our results uncover race-related investment choices at both levels.
Han, Lina, Xing Huang, Ohad Kadan, and Jimmy Wu. 2026. “The True Colors of Money: Racial Representation and Asset Management.” Review of Financial Studies. https://doi.org/10.1093/rfs/hhag038.
```
@article{Han_Huang_Kadan_Wu_2026_RFS,
  author  = {Han, Lina and Huang, Xing and Kadan, Ohad and Wu, Jimmy},
  title   = {The True Colors of Money: Racial Representation and Asset Management},
  journal = {Review of Financial Studies},
  year    = {2026},
  doi     = {10.1093/rfs/hhag038}
}
```

Working Papers

Instruction Tuning Chronologically Consistent Language Models

with Songrun He, Linying Lv, and Asaf Manela

Working paper
Abstract

We introduce a family of chronologically consistent, instruction-following large language models to eliminate lookahead bias. Each model is trained only on data available before a clearly defined knowledge-cutoff date, ensuring strict temporal separation from any post-cutoff data. The resulting framework offers (i) a simple, conversational chat interface, (ii) fully open, fixed model weights that guarantee replicability, and (iii) a conservative lower bound on forecast accuracy, isolating the share of predictability that survives once training leakage is removed. Together, these features provide researchers with an easy-to-use generative AI tool useful for a wide range of prediction tasks that is free of lookahead bias.
He, Songrun, Linying Lv, Asaf Manela, and Jimmy Wu. 2025. “Instruction Tuning Chronologically Consistent Language Models.” Working paper. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5348747.
```
@unpublished{He_Lv_Manela_Wu_2025_InstructionTuning,
  author = {He, Songrun and Lv, Linying and Manela, Asaf and Wu, Jimmy},
  title  = {Instruction Tuning Chronologically Consistent Language Models},
  year   = {2025},
  note   = {Working paper},
  url    = {https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5348747}
}
```
News Perplexity

Solo-authored

Working paper
Abstract

Existing research suggests that investors inefficiently process text that induces cognitive strain. This paper challenges that view. Drawing on dual-process theory, I hypothesize that certain strain-inducing text disrupts heuristic (System 1) intuition and triggers deliberate (System 2) analysis. To capture such text, I introduce news perplexity—a metric derived from large language models—and find that high perplexity news significantly mitigates market underreaction. Validating the mechanism, high perplexity news coincides with a spike in EDGAR search traffic, which is consistent with the notion that investors respond to the cognitive trigger by acquiring more information. These findings provide the first evidence that beneficial cognitive strain can enhance market efficiency by forcing deeper information processing.
Wu, Jimmy. 2025. “News Perplexity.” Working paper.
```
@unpublished{Wu_2025_NewsPerplexity,
  author = {Wu, Jimmy},
  title  = {News Perplexity},
  year   = {2025},
  note   = {Working paper}
}
```

Jimmy Wu

About

Research

Publications

Working Papers

Teaching

Instructor · University of Iowa

Education

Cite this paper

ChronoBERT

ChronoGPT

ChronoGPT-instruct