ECS579 Transforming Economic and Financial Research with AI

• Topics

  This doctoral course provides an introduction to how recent advances in generative AI are reshaping research in economics and finance. The course covers how to integrate AI into research workflows, discusses applications from recent research papers, introduces tools for working with large language models at scale, and explores how participants can identify opportunities to apply AI in their own research projects. The course will also cover recent developments in agentic coding, including the use of coding agents and reproducible AI workflows for research.

• Learning outcome

  Upon completion of the course, the student can:

  Knowledge

  • Explain how recent advances in generative AI are transforming research practices in economics and finance.
  • Discuss key applications of large language models in economic and financial research, including qualitative interviews, behavioral interventions, and large-scale text analysis.
  • Evaluate the opportunities and limitations of agentic coding, coding agents, and reproducible AI workflows for research.

  Skills

  • Apply generative AI tools to support large-scale data analysis and reproducible research workflows.
  • Assess the advantages and limitations of integrating AI into different stages of research design.
  • Design AI-supported workflows with clear procedures for quality assurance and validation.
  • Use tools for working with large language models at scale, including coding agents and version-controlled workflows.

  General competence

  • Reflect critically on the methodological and practical implications of using AI in economic and financial research.

• Teaching

  The teaching consists of ordinary lectures, guest lectures, hands-on sessions with Codex, and student presentations. 

• Restricted access

  • PhD candidates at NHH
  • PhD candidates at Norwegian institutions
  • PhD candidates at other institutions.
  • PhD candidates from the ENGAGE.EU alliance.
  • Motivated master’s students may be admitted after application, but are subject to the approval from thecourse responsible on a case by case basis

• Recommended prerequisites

  None. 

• Compulsory Activity

  Slides outlining motivation and plan for the term paper, submitted in groups of 3 to 4 students.

  The compulsory activity (work requirement) is only valid for one semester. 

• Assessment

  A group term paper on exploiting generative AI at scale. The paper must be written in English, and completed in groups of 3-4 students. The paper is due four weeks after course completion.

• Grading Scale

  Pass-Fail

• Computer tools

  Students are strongly recommended to install the Codex CLI and Git before the course starts.

• Literature

  Research papers

  Chopra, F., & Haaland, I. (2023). Conducting qualitative interviews with AI. https://github.com/fchop/papers/raw/gh-pages/Qualitative%20Interviews.pdfhttps://github.com/fchop/papers/raw/gh-pages/Qualitative%20Interviews.pdf

  Chopra, F., Haaland, I., Roever, N., & Roth, C. (2026). Evaluating behavioral interventions at scale with AI. https://www.econtribute.de/RePEc/ajk/ajkdps/ECONtribute_385_2026.pdfhttps://www.econtribute.de/RePEc/ajk/ajkdps/ECONtribute_385_2026.pdf

  Chopra, F., Haaland, I., Roeben, F., Roth, C., & Sticher, V. (2025). News customization with AI. https://www.econstor.eu/bitstream/10419/331587/1/cesifo1_wp12121.pdfhttps://www.econstor.eu/bitstream/10419/331587/1/cesifo1_wp12121.pdf

  Braghieri, L., Eichmeyer, S., Levy, R., Mobius, M., Steinhardt, J., & Zhong, R. (2024). Article-level slant and polarization of news consumption on social media. CEPR Discussion Paper No. 19807. https://cepr.org/publications/dp19807?utm_source=chatgpt.comhttps://cepr.org/publications/dp19807

  Software and documentation

  OpenAI. (2026). Codex CLI documentation. https://developers.openai.com/codex/clihttps://developers.openai.com/codex/cli

  Git. (2026). Git downloads. https://git-scm.com/downloadshttps://git-scm.com/downloads