ID:
APFAI26
Duration (hours):
36
CFU:
6
SSD:
Economia applicata
Located in:
PESCARA
Url:
MANAGEMENT AND ECONOMICS/FINANZA, BANCHE E MERCATI Year: 1
Year:
2026
Course Catalogue:
Overview
Date/time interval
Secondo Quadrimestre (06/01/2027 - 14/04/2027)
Syllabus
Course Objectives
The course aims to train graduates capable of critically assessing the adoption of artificial intelligence systems by financial intermediaries and markets, combining technical understanding, economic analysis and regulatory awareness.
Knowledge and understanding. Students know the taxonomy of AI and the operating logic of the main machine learning models; they know the use cases of AI in the financial sector and the relevant European regulatory framework.
Applying knowledge and understanding. Students can analyse a real use case by identifying the prediction produced, the data feeding it, the role of human judgement and the risks generated; they can interpret the output of a predictive model and reason about the trade-off between classification errors in economic terms.
Making judgements. Students critically assess claims about the performance of AI systems, distinguishing empirical evidence from commercial rhetoric, and recognise the methodological limits of adoption surveys; they can argue about the trade-offs between accuracy, interpretability and fairness.
Communication skills. Students present, in written and oral form, the analysis of an AI use case to both technical and non-technical audiences, justifying their assessments.
Learning skills. Students are able to independently read academic papers and supervisory documents in a rapidly evolving field, updating their knowledge beyond the end of the course.
Knowledge and understanding. Students know the taxonomy of AI and the operating logic of the main machine learning models; they know the use cases of AI in the financial sector and the relevant European regulatory framework.
Applying knowledge and understanding. Students can analyse a real use case by identifying the prediction produced, the data feeding it, the role of human judgement and the risks generated; they can interpret the output of a predictive model and reason about the trade-off between classification errors in economic terms.
Making judgements. Students critically assess claims about the performance of AI systems, distinguishing empirical evidence from commercial rhetoric, and recognise the methodological limits of adoption surveys; they can argue about the trade-offs between accuracy, interpretability and fairness.
Communication skills. Students present, in written and oral form, the analysis of an AI use case to both technical and non-technical audiences, justifying their assessments.
Learning skills. Students are able to independently read academic papers and supervisory documents in a rapidly evolving field, updating their knowledge beyond the end of the course.
Course Prerequisites
Students are expected to have the basic knowledge acquired in an undergraduate degree in economics and management: elements of statistics and econometrics (random variables, estimation, linear regression, hypothesis testing) and the fundamentals of financial intermediation and corporate finance (how banks and markets work, risk and return, valuation of financial instruments).
No programming skills are required: the course labs use interactive tools prepared by the instructor, usable from any browser without installation or coding. A reading knowledge of English sufficient for technical documents is useful but not required.
No programming skills are required: the course labs use interactive tools prepared by the instructor, usable from any browser without installation or coding. A reading knowledge of English sufficient for technical documents is useful but not required.
Teaching Methods
The course consists of 36 contact hours, organised in 12 three-hour lectures, and combines three modes of work.
Interactive lectures. Each lecture starts from the questions raised by the week's lecture notes and alternates theoretical exposition with guided case discussion.
Hands-on labs. Each lecture includes a lab (approximately 45-60 minutes) run in class on standalone web applications prepared by the instructor: scoring simulators, model explorers, market simulations, document analysis exercises. No programming is required. Labs can be reused at home and end with a submission sheet.
Work on primary sources and group project. Attending students work in groups of 3-4 on the critical analysis of a real AI use case in a financial intermediary or market, presented and discussed in the final lecture. The topic is chosen by the group, subject to the instructor's approval.
Teaching materials (lecture notes, slides, labs) are published on the university platform after each lecture.
Interactive lectures. Each lecture starts from the questions raised by the week's lecture notes and alternates theoretical exposition with guided case discussion.
Hands-on labs. Each lecture includes a lab (approximately 45-60 minutes) run in class on standalone web applications prepared by the instructor: scoring simulators, model explorers, market simulations, document analysis exercises. No programming is required. Labs can be reused at home and end with a submission sheet.
Work on primary sources and group project. Attending students work in groups of 3-4 on the critical analysis of a real AI use case in a financial intermediary or market, presented and discussed in the final lecture. The topic is chosen by the group, subject to the instructor's approval.
Teaching materials (lecture notes, slides, labs) are published on the university platform after each lecture.
Assessment Methods
Attending students. Assessment is distributed across four components: two written mid-term tests taken in class (25% each), lasting 45 minutes, consisting of closed-answer questions and one short open question, covering Lectures 1-4 and Lectures 5-8 respectively; a group project (40%), consisting of the critical analysis of a real AI use case in finance, presented and discussed in the final lecture; participation in the labs (10%), verified through the submission of at least 6 lab sheets out of 10. Students who do not pass a mid-term test, or who wish to improve their result, may take the corresponding part in the final examination.
The group project is assessed according to a rubric communicated at the beginning of the course: technical understanding of the system analysed (25%), economic analysis of value, incentives and effects on customers and markets (30%), analysis of risks and regulatory compliance (25%), quality of the presentation and ability to answer questions (20%).
Non-attending students. Assessment consists of a final written examination on the whole programme (60%), with closed-answer questions and two open questions, and an individual paper (40%) of approximately 3,000 words analysing an AI use case in finance, discussed orally. The topic of the paper must be agreed with the instructor at least three weeks before the examination date.
Assessment criteria. In both paths the final mark, expressed out of thirty, reflects: command of concepts and technical vocabulary, ability to apply the course's analytical framework to new cases, rigour of the economic and regulatory argument, and clarity of exposition. Honours (lode) are awarded where independence of judgement and the ability to connect the parts of the programme are particularly strong.
The group project is assessed according to a rubric communicated at the beginning of the course: technical understanding of the system analysed (25%), economic analysis of value, incentives and effects on customers and markets (30%), analysis of risks and regulatory compliance (25%), quality of the presentation and ability to answer questions (20%).
Non-attending students. Assessment consists of a final written examination on the whole programme (60%), with closed-answer questions and two open questions, and an individual paper (40%) of approximately 3,000 words analysing an AI use case in finance, discussed orally. The topic of the paper must be agreed with the instructor at least three weeks before the examination date.
Assessment criteria. In both paths the final mark, expressed out of thirty, reflects: command of concepts and technical vocabulary, ability to apply the course's analytical framework to new cases, rigour of the economic and regulatory argument, and clarity of exposition. Honours (lode) are awarded where independence of judgement and the ability to connect the parts of the programme are particularly strong.
Texts
No single textbook is adopted. Examinable material consists of the instructor's lecture notes (one per lecture, distributed via the university platform) and the primary readings assigned lecture by lecture, whose main items are listed below.
Reports by authorities:
- Financial Stability Board (2024), The Financial Stability Implications of Artificial Intelligence.
- Financial Stability Board (2025), Monitoring Adoption of Artificial Intelligence and Related Vulnerabilities in the Financial Sector.
- Bank of England (2025), Financial Stability in Focus: Artificial Intelligence in the Financial System.
- Bank for International Settlements (2025), The Use of Artificial Intelligence for Policy Purposes.
- Perez-Cruz F., Prenio J., Restoy F., Yong J. (2025), Managing Explanations: How Regulators Can Address AI Explainability, BIS Occasional Paper no. 24.
- European Banking Authority (2025), AI Act: Implications for the EU Banking and Payments Sector.
- ESMA, Public Statement on the Use of Artificial Intelligence in the Provision of Retail Investment Services.
Academic literature:
- Mullainathan S., Spiess J. (2017), "Machine Learning: An Applied Econometric Approach", Journal of Economic Perspectives, 31(2).
- Fuster A., Goldsmith-Pinkham P., Ramadorai T., Walther A. (2022), "Predictably Unequal? The Effects of Machine Learning on Credit Markets", Journal of Finance, 77(1).
- Gu S., Kelly B., Xiu D. (2020), "Empirical Asset Pricing via Machine Learning", Review of Financial Studies, 33(5) (selected sections indicated in class).
- Foucault T., Gambacorta L., Jiang W., Vives X. (2025), "Artificial Intelligence in Finance", VoxEU.
All readings are available in open access or through the university's electronic resources. The final, updated list is published at the beginning of the course.
Reports by authorities:
- Financial Stability Board (2024), The Financial Stability Implications of Artificial Intelligence.
- Financial Stability Board (2025), Monitoring Adoption of Artificial Intelligence and Related Vulnerabilities in the Financial Sector.
- Bank of England (2025), Financial Stability in Focus: Artificial Intelligence in the Financial System.
- Bank for International Settlements (2025), The Use of Artificial Intelligence for Policy Purposes.
- Perez-Cruz F., Prenio J., Restoy F., Yong J. (2025), Managing Explanations: How Regulators Can Address AI Explainability, BIS Occasional Paper no. 24.
- European Banking Authority (2025), AI Act: Implications for the EU Banking and Payments Sector.
- ESMA, Public Statement on the Use of Artificial Intelligence in the Provision of Retail Investment Services.
Academic literature:
- Mullainathan S., Spiess J. (2017), "Machine Learning: An Applied Econometric Approach", Journal of Economic Perspectives, 31(2).
- Fuster A., Goldsmith-Pinkham P., Ramadorai T., Walther A. (2022), "Predictably Unequal? The Effects of Machine Learning on Credit Markets", Journal of Finance, 77(1).
- Gu S., Kelly B., Xiu D. (2020), "Empirical Asset Pricing via Machine Learning", Review of Financial Studies, 33(5) (selected sections indicated in class).
- Foucault T., Gambacorta L., Jiang W., Vives X. (2025), "Artificial Intelligence in Finance", VoxEU.
All readings are available in open access or through the university's electronic resources. The final, updated list is published at the beginning of the course.
Contents
The course examines the impact of artificial intelligence on the financial industry from an applied economics perspective. The approach is not computational but economic: AI is treated as a technology that lowers the cost of prediction, and the guiding question is which tasks it makes cheaper, who captures the resulting value, which risks it generates for customers, intermediaries and systemic stability, and how regulation governs them.
The course is organised in four parts. The first builds the conceptual foundations required to critically assess an AI system: taxonomy (machine learning, deep learning, generative AI), the logic of supervised learning, model evaluation criteria, interpretability. The second covers the core applications in finance: creditworthiness assessment, return prediction and asset pricing, algorithmic trading and market microstructure, portfolio management and robo-advisory, fraud detection and anti-money laundering. The third is devoted to generative AI and large language models and to their integration into banking and insurance processes. The fourth addresses the European regulatory and prudential framework — the AI Act, DORA, supervision by EBA/ESMA/EIOPA — and the implications for financial stability.
Each lecture is paired with an interactive lab that requires no programming skills.
The course is organised in four parts. The first builds the conceptual foundations required to critically assess an AI system: taxonomy (machine learning, deep learning, generative AI), the logic of supervised learning, model evaluation criteria, interpretability. The second covers the core applications in finance: creditworthiness assessment, return prediction and asset pricing, algorithmic trading and market microstructure, portfolio management and robo-advisory, fraud detection and anti-money laundering. The third is devoted to generative AI and large language models and to their integration into banking and insurance processes. The fourth addresses the European regulatory and prudential framework — the AI Act, DORA, supervision by EBA/ESMA/EIOPA — and the implications for financial stability.
Each lecture is paired with an interactive lab that requires no programming skills.
Course Language
Italian.
Course materials (lecture notes, slides, labs) are in Italian. A substantial part of the assigned readings — academic papers and reports issued by international supervisory authorities — is in English: for each reading the lecture notes provide guidance and a summary in Italian. The examination is held in Italian; upon the student's request it may be taken in English.
Course materials (lecture notes, slides, labs) are in Italian. A substantial part of the assigned readings — academic papers and reports issued by international supervisory authorities — is in English: for each reading the lecture notes provide guidance and a summary in Italian. The examination is held in Italian; upon the student's request it may be taken in English.
More information
Attendance. Attendance is not compulsory, but the course provides distinct assessment paths for attending and non-attending students. Students are considered attending if they take part in at least 75% of the lectures (9 out of 12) and meet the lab submission requirements.
Use of artificial intelligence tools. The use of generative AI tools for study purposes is allowed and encouraged, consistently with the subject of the course. In assessments and in the group project, any use of such tools must be declared, specifying at which stage of the work and for what purpose: declaring it does not penalise the assessment, while undeclared use does. University rules on academic integrity in examinations continue to apply.
Students with disabilities or specific learning disorders. Students with disabilities or specific learning disorders may agree with the instructor, in coordination with the university services, on the compensatory tools and dispensatory measures provided for by law.
Office hours. By appointment, in person or online, to be arranged by email.
Use of artificial intelligence tools. The use of generative AI tools for study purposes is allowed and encouraged, consistently with the subject of the course. In assessments and in the group project, any use of such tools must be declared, specifying at which stage of the work and for what purpose: declaring it does not penalise the assessment, while undeclared use does. University rules on academic integrity in examinations continue to apply.
Students with disabilities or specific learning disorders. Students with disabilities or specific learning disorders may agree with the instructor, in coordination with the university services, on the compensatory tools and dispensatory measures provided for by law.
Office hours. By appointment, in person or online, to be arranged by email.
Degrees
Degrees
MANAGEMENT AND ECONOMICS
Master’s Degree
2 years
No Results Found