Dr Felipe Costa Sperb

Dr Sperb has extensive experience in lecturing, as well as designing and managing courses for both undergraduate and postgraduate programmes. These include courses in management, finance, management science, research methods, business and data analytics, delivered within both Business School and School of Informatics settings.

Dr Sperb is also highly experienced in supervising undergraduate and postgraduate dissertations across a wide range of programmes, including AI, Data Science, Computer Science, Financial Computing, Finance, Management, Banking, Risk, and Operational Research. Additionally, Dr Sperb has supervised PhD research in the areas of AI, machine learning and prediction analytics applied to financial and corporate decision-making and risk management.

Courses taught

• Business Management BSc (Hons)

Research Interests

Dr Sperb's research focuses on Risk Analysis, AI and Predictive Modelling, and Behavioural and Decision Analytics with multidisciplinary data-driven approaches in naturalistic financial and business settings.

Articles

• Costa Sperb, L. F., Sung, M.S., Ma, T., and Johnson, J.E.V. (2022). Turning the heat on financial decisions: Examining the role temperature plays in the incidence of bias in a time-limited financial market, European Journal of Operational Research, 299(3), p. 1142-1157.
• Costa Sperb, L. F., Sung, M.S., Ma, T., and Johnson, J.E.V. (2019). Keeping a weather eye on prediction markets: The influence of environmental conditions on forecasting accuracy, International Journal of Forecasting, 35(1), p. 321-335.

Discussion and Working Papers

• Xu, V., Li, W., Leang, J., Ma, T., Costa Sperb, L. F., He., F. (2026). Explainable self-attention modelling for financial vulnerability prediction, Edinburgh Centre for Financial Innovations, Edinburgh, UK.
• Robertson-Rose, L., Costa Sperb, L. F., Ma, T., Majid, H. and Tugay, R. (2022). Older workers, later lives: Financial vulnerability amongst the over 50s, Smart Data Foundry. Edinburgh, UK.
• Costa Sperb, L.F. and Marshall, A. (2020). The risk radar and the view of risk: A Centre for Risk Research discussion booklet, Centre for Risk Research. Southampton, UK.
• Ravishankar, V., Mooney, O., Hader, N., Costa Sperb, L. F., Hempson-Jones, J. (2018). Stepping up governance on cyber security: What is corporate disclosure telling investors, Principles for Responsible Investment. London, UK.
• Costa Sperb, L. F., Sung, M.S., Ma, T., and Johnson, J.E.V. (2017). Determining preference for weather and atmospheric conditions: Forecasting methodology and an application, University of Southampton Working Paper. Southampton, UK.  

Conference or Workshop Item

• Hao, Z., Costa Sperb, L. F., Ma, T. (2025). Zero-shot extraction of stock relationship graphs with LLMs. In Proceedings of The 10th Workshop on Financial Technology and Natural Language Processing, pages 12–18, Suzhou, China. Association for Computational Linguistics.
• Costa Sperb, L. F. and Ma, T. (2023). Can online news influence our risk attitudes and decisions? Evidence from a real-world financial market. In: Economics of Financial Technology Conference; Edinburgh, UK.
• Zhou, H., Costa Sperb, L. F. and Ma, T. (2023). Toward a multiple dimensional risk factors and temporal dependencies transformer model for stock price movement prediction. In: International Conference on AI in Finance; New York, USA.
• Costa Sperb, L. F. (2023). Textual sentiment, investor behaviour, and financial asset pricing. In: Artificial Intelligence and its Applications Institute Seminar; Edinburgh, UK.
• Costa Sperb, L. F. (2023). Course design principles for interdisciplinary taught postgraduate programmes. In: University of Edinburgh Teaching Seminar; Edinburgh, UK.
• Costa Sperb, L. F. (2022). Roles and profiles of supervisors in the context of practical dissertation projects. In: University of Edinburgh Teaching Festival; Edinburgh, UK.
• Costa Sperb, L. F. (2022). Teaching research methods: How to make it meaningful to students from different disciplines. In: University of Edinburgh Teaching Festival; Edinburgh, UK.
• Costa Sperb, L. F. and Ma, T. (2021). Using sentiment from online news to quantify risks that happen in the global economic space. In: International Conference on AI in Finance; New York, USA.
• Costa Sperb, L. F. and Ma, T. and (2020). Reading between the lines: Extracting cyber risk information from news sentiment. In: Workshop on Emerging Cyber-Risk Challenges for Finance, Insurance, and Technology Sectors: How Academic and Industry Leaders Effectively Work Together; Edinburgh, UK.
• Ma, T. and Costa Sperb, L. F. (2020). Challenges with quantifying cyber risk from an academic perspective. In: Cyber Risk and Finance Industry Symposium by the National Cyber Security Centre and the Institute and Faculty of Actuaries; London, UK.
• Costa Sperb, L. F., Marshall, A., Roussou, F., and Mues, C. (2018). Risk radar: A methodology to categorise risks disclosed on SEC 10-K reports. In: Workshop on Risk Innovation Solutions by Munich Re Group, Institute of Risk Management, and the Centre for Risk Research (University of Southampton); Southampton, UK.
• Costa Sperb, L. F., Sung, M.S., Ma, T., and Johnson, J.E.V. (2017). Smart(er) money: Using environmental conditions information to enhance statistical and economic value of probabilistic forecasts. In: Southampton Business School Research Seminar; Southampton, UK.
• Costa Sperb, L. F., Sung, M.S., Ma, T., and Johnson, J.E.V. (2016). Towards a better understanding of the influence of weather on behaviour and the extent to which this is understood by financial markets. In: Southampton Business School Summit on Management and Business Analytics; Southampton, UK.