Research: UH Hilo Assistant Professor of Finance Wei Wei on AI, LLMs, NLP, and machine learning
Assistant Professor of Finance Wei Wei’s research focuses on financial technology, corporate finance, and financial accounting, all with a healthy dose of AI application and machine learning.
By Susan Enright/UH Hilo Stories.
This story on Wei Wei’s research profile was first published on the website Keaohou that features UH Hilo faculty research and scholarly activity.
A new finance professor at the University of Hawaiʻi at Hilo is focusing much of her research on the application of artificial intelligence and machine learning to financial analysis.
Wei Wei is an assistant professor of finance with research interests in financial technology, corporate finance, and financial accounting. Notable in her investigations is not only the pursuit of learning more about the application of AI and machine learning to financial analysis, but also applying these technologies to her own investigations.
Wei received her master of business in finance from the University of Minnesota’s Carlson School of Management, and her doctoral degree in finance from University of North Carolina at Charlotte. She is a certified Financial Risk Manager (FRM) and pursued the Chartered Financial Analyst (CFA) program, strengthening her expertise in risk management and investments.
Before coming to UH Hilo in 2025, she was a lecturer at UNC’s Charlotte Belk College of Business. Now at UH Hilo, Wei appreciates the university’s unique setting that allows her to connect business education with the cultural and social context of Hawaiʻi.
While Wei’s previous publications include investigations into audits of Chinese accounting firms and inadequate disclosures practices in relation to erosion in shareholder wealth, among her current interests is inquiry into how markets process information and how large language models, or LLMs, can detect fraud and predict market reactions. She is researching the application of AI and machine learning to financial analysis — “specifically, improving how we detect risks and extract insights from complex financial data,” she says.
Working papers
Assistant Professor Wei has several current working papers underway, two under review with prestigious journals.
Title: “How to write returns: a counterfactual editor for earnings guidance and press releases”
One investigation takes a look contextually at the way more and more investors read and analyze text with LLMs. “This paper moves beyond predictive text analytics to a prescriptive question: how should managers write earnings disclosures to improve investor understanding?”, says Wei in her abstract.
“I develop a series of counterfactual CAR editor that proposes fact-preserving edits to earnings press releases and guidance updates, and reports the predicted change in abnormal returns (CAR) using widely-used LLMs including ChatGPT, Claude, and Gemini with uniform prompts.”
“The results prove LLMs can better capture the underlying information from text by fact-preserving edits,” she concludes.
Title: “Audio context: the causal effect of CEO communication”
In other research, Wei examines chief executive officer communication, noting there is evidence that CEOs have particular management styles that affect firm performance. On the other hand, Wei writes, CEO-firm matches are highly endogenous in that boards hire CEOs with management styles that best match the firm’s needs.
With co-author Patrick S. Smith, the two researchers examine how CEOs’ ability to communicate their style, not the style itself, affects analyst forecasts and future firm performance.
“Implicit in our analysis is the idea that CEOs can have similar styles yet produce different results,” says Wei. “Using Natural Language Processing (NLP), we abstract CEOs’ arousal level from Q&A session of earnings calls, and find communication is one channel through which CEOs have a causal effect on the firms they run.”
Title: “Racial diversity and firm commitments”
A paper currently under review with the journal Accounting and Business Research provides valuable insights into the socio-economic drivers behind corporate diversity policies. With co-author Al (Aloke) Ghosh, Wei analyzes key factors driving corporate commitment toward enhancing racial diversity following the death of George Floyd and their effects on diversity performance.
“Using textual analysis of earnings calls and 10-K reports (corporate communications), my findings reveal a significant increase in race-related discussions for the post-event period,” says Wei. “Key drivers of the increase in racial diversity discussions in corporate communications include corporate integrity, strong ESG performance [an investing principle that prioritizes environmental issues, social issues, and corporate governance], and the strength of pre-existing diversity policies.”
In this study, Wei also examines whether elevated corporate diversity communications coincide with efforts to improve DEI commitments, or whether firms engage in “diversity washing.”
“Based on a difference-in-differences research design, firms discussing race-related matters demonstrate a marked improvement in DEI performance metrics,” says Wei. “Further, my results indicate that analysts and rating agencies view these changes favorably, and some evidence that investors respond positively to DEI improvements.”
Title: “Short interest and financial reporting misstatements”
In another study also currently under review, this one with the Journal of Finance Reporting, Wei investigates short interest and financial reporting misstatements. This work explores a rare interest rate phenomenon, called an inverted U-shaped short interest curve, when coupled with announcements of misstatements; prior studies conclude that short sellers can predict financial reporting misstatements.
“My re-examination using a comprehensive sample fails to find convincing evidence of a pronounced inverted U-shaped curve around most reporting misstatements,” says Wei. “However, I find heightened short-selling activity both before and after misstatement announcements.”
In her testing of whether this result is attributable to short sellers’ superior ability to analyze the release of fundamental adverse corporate news, she finds most large increases in short interest levels leading up to misstatement announcements “coincide with the disclosures of poor performance, legal troubles, stock downgrade, potential delisting, impending bankruptcy, stock sale, departure of key employees, the closure of businesses, and greater financial challenges.”
“I infer that the elevated short interest before the announcement of misstatements is linked to short sellers’ ability to process public information and is not based entirely on non-public information,” she concludes.
Instilling curiosity in her students along with a strong sense of ethics and social justice
Wei says she actively integrates her empirical research into her teaching, both to expose students to cutting-edge tools and methods and to involve them directly in the research process.
“This gives students hands-on experience with real data, analytical tools, and collaborative problem-solving, helping prepare them for careers in finance, analytics, and related fields,” she says.
She also inspires her students to pursue this line of inquiry into their own research projects and internships.
“This type of work helps automate routine analysis while enhancing accuracy and transparency,” she says. “For graduates, understanding how AI can support financial decision-making is becoming an essential skill. Employers increasingly look for people who can interpret data, work with analytical tools, and use technology responsibly.”
Wei says this includes those in finance having a strong sense of ethics and social justice, values she instills in her students.
Fundamentally, through learning and scientific inquiry in both the classroom and through internships, Wei hopes her students will develop the essential skills of critical thinking, resilience, and ethical awareness — along with the belief that they can master complex topics and use financial knowledge to make a positive difference in their communities.
Story by Susan Enright, a public information specialist for the Office of the Chancellor and editor of UH Hilo Stories. She received her bachelor of arts in English and certificate in women’s studies from UH Hilo.
