My Research Philosophy

My research sits at the intersection of natural language processing, large language models, and empirical corporate finance. I'm interested in how text, whether it's SEC filings, analyst reports, local news coverage, or patent documents, encodes information that markets and firms don't always price or act on correctly. Much of my work builds computational tools (LLM-based classifiers, fine-tuned sentiment models, large-scale NLP pipelines) to extract and validate signals from unstructured financial text at a scale that wasn't previously possible, then tests whether those signals matter for real outcomes: crash risk, disclosure quality, capital allocation, or the erosion of local financial information ecosystems. I care about work that is methodologically rigorous but also legible and useful to practitioners, regulators, and the broader accounting and finance community, not just to other academics

Published Work

Aguegboh, E. S., Onuoha, U. C., & Patel, P. (2026). Assessing the relevance of sell-side analyst recommendations. Review of Financial Economics, 44, e70015. https://doi.org/10.1002/rfe.70015

Kumar, S., Chakraborty, A. ‘., & Patel, P. (Aug 2026). Prompt Engineering for Accounting and Finance (p. 736). Springer Nature Switzerland. https://doi.org/10.1007/978-3-032-11195-1

Working Papers

The Price of a News Desert: The Impact of Newspaper Closures on Financial Service Quality

Co-authors: John (Jianqiu) Bai, Chi Wan, Donghua Zhou.

Examines how the closure of local newspapers affects the quality of financial services in the communities they served. Presented at CICF 2026 and the MRS International Risk Conference (2025).

Beyond AI Coding: Teaching Human Judgment in Finance Education

Co-authors: Atreya Chakraborty, Honggang Qiu
Explores how finance education can be designed to cultivate human judgment alongside growing reliance on AI tools in coding and analysis.

Dosage Matters: A Framework for Calibrated AI Integration Across the Learning Trajectory

Co-authors: Sunil Kumar, Honggang Qiu
Proposes a framework for calibrating the integration of AI tools across different stages of the learning process.

Reading Between the Lines: Contextual Sentiment Gaps and Crash Risk in Corporate Disclosure

Co-authors: Joshua Burke, Atreya Chakraborty, Honggang Qiu
Studies the gap between contextual and literal sentiment in corporate disclosures and its relationship to stock price crash risk.

Climate Regulation Risk Exposure and the Innovation Response of Firms

Co-authors: Li Ai, Lucia Silva Gao
Examines how firms' exposure to climate regulation risk shapes their innovation activity. Presented at the Southern Finance Association Annual Conference, FMA Europe Annual Conference, and Global Finance Conference (2026).

When More Means Less: Economic Policy Uncertainty and the Informational Quality of Forward-Looking Disclosure

Co-authors: Ekene S. Aguegboh, Uchenna C. Onuoha
Studies how economic policy uncertainty affects the quality and informativeness of firms' forward-looking disclosures. Presented at the 3rd Modern Finance Conference.

Working Paper Click here !

Invented Here, Owned Elsewhere? Organizational Fragmentation, Patent Ownership and Territorial Value Capture Across U.S. Counties

Co-authors: Sébastien Bourdin, Atreya Chakraborty
Investigates how organizational fragmentation affects where patent ownership is held versus where innovation actually happens, using county-level data across the U.S.

Circular Economy: Using LLM for Circular Patent Classification

Co-authors: Sébastien Bourdin, Atreya Chakraborty, Julian Kirchherr
Develops a large language model-based framework to classify patents according to circular economy principles.