Research & publications

This page is a short overview. For a dated, complete record (peer-reviewed papers, software, teaching publications, and CV downloads), use the résumé site and ORCID.

My work sits where finance, statistics, and trustworthy machine learning meet: markets and institutions, risk, and evidence that can stand up in operational and regulatory settings. I collaborate across disciplines and with industry, and I try to keep assumptions explicit and claims proportionate to the data.

Research labs

  • Centre for Finance and Responsible Technology (CFRT), Ulster University Business School. Current research home, where I serve as Director.
  • Finance and AI Research (FAIR) lab, Queen’s University Belfast. Developed before my move to Ulster.

Methodological themes

Three strands recur across my research programme:

  1. Trustworthy machine learning for market integrity and risk: representation learning for detecting market manipulation (self-supervised frameworks, signal amplification for rare events, transformer-based sentiment models for crash-risk prediction). The methodological commitment is that model performance and model trustworthiness are separable properties, and both must be evidenced before deployment in operational or regulatory settings.

  2. Computer vision for socio-economic measurement: using vision models on real-world imagery to produce auditable indicators (geospatial vehicle-crime modeling with street-level features; hierarchical flag classification for cultural-symbol monitoring). These pipelines are treated as measurement instruments and evaluated on measurement-science grounds, not on benchmark accuracy alone.

  3. Causal and statistical evidence for governance and regulation: innovation signals and strategic exits in computational approaches to financial regulation; ownership dynamics, risk, and regulation in Chinese banking; collaborative doctoral partnerships applying causal inference and econometrics to live regulatory and compliance problems. The priority is that evidence survive operational and regulatory scrutiny: assumptions explicit, uncertainty quantified, and claims proportionate to what the data can support.

Selected peer-reviewed work

A few representative papers (venues and links as on the résumé; the résumé list is authoritative if anything differs):

  • Wang, Dai, Spence, Rafferty, Quinn & Huang (2025). TDSRL: Time series dual self-supervised representation learning for anomaly detection from different perspectives. IEEE Internet of Things Journal.
  • Liu et al. (2024). Evolutionary multi-objective optimisation for large-scale portfolio selection with both random and uncertain returns. IEEE Transactions on Evolutionary Computation (ABS4).
  • Quinn, Gallagher & Kuosmanen (2023). Lurking in the shadows: The impact of CO₂ emissions target setting on carbon pricing in the Kyoto agreement period. Energy Economics (ABS3).
  • McKillop, Liu, Quinn & Peng (forthcoming). Modelling and predicting credit union failures. International Journal of Forecasting (ABS3).
  • Bouri, Quinn, Sheenan & Tang (2024). Investigating extreme linkage topology in the aerospace and defence industry. International Review of Financial Analysis (ABS3).
  • Liu, Papailias & Quinn (2021). Direction-of-change forecasting in commodity futures markets. International Review of Financial Analysis (ABS3).
  • Gallagher & Quinn (2020). Regulatory own goals: The unintended consequences of economic regulation in professional football. European Sport Management Quarterly (ABS3).

Full list on the résumé.

Working papers and preprints

Active pipeline (status as known; see résumé for latest):

  • Birem, Abidi Perier, Quinn & Kearney. Transformer-based sentiment analysis for stock market crash risk. Working paper.
  • Dai, Quinn, Kearney & Wang. Amplifying market manipulation detection signals. Working paper.
  • Dai, Quinn, Kearney, Liu, Spence, Rafferty & Wang. Detecting market manipulation with dual-branch self-supervised learning: A unified framework integrating frequency-informed anomaly synthesis and domain-specific features. Working paper.
  • Zhang, Quinn & Sheenan (2025). Ownership dynamics, risk and regulation in Chinese banking: New evidence. Centre for Responsible Banking & Finance Working Paper Series, WP No. 25-017.
  • Kearney, Quinn & Pramanick (under review). Innovation signals and strategic exits: How technological readiness shapes outcomes in computational approaches to financial regulation. Journal of Business Venturing.
  • Hannon, French, Quinn & O’Hagan (submitted). Geospatial modeling of vehicle crime in Northern Ireland using computer vision to identify environmental factors. Insurance: Mathematics and Economics (IME-D-25-00419).
  • Quinn (2025). Hierarchical flag classification through economic domain knowledge: A vision transformer approach for cultural symbol recognition. MSc thesis and reproducible research codebase (economic-flag-classification); manuscript for journal submission in preparation.
  • Quinn & Gallagher. Great expectations: Managerial turnover and market expectation in association football. Working paper.
  • Quinn (2023). Explaining AI in finance: Past, present, prospects. arXiv:2306.02773.
  • Quinn (2022). Teaching open science analytics in the age of financial technology. SSRN.

Funded research and knowledge transfer

Active and representative funded work (amounts and dates as recorded at award; updated detail on the résumé):

  • Three PhD scholarships to support the Centre for Finance and Responsible Technology, including two collaborative doctoral partnerships with Pytillia and Napier AI (Department of the Economy NI, £360k, 2025–).
  • Understanding and enhancing regulatory compliance in the investment management industry using AI, with Funds Axis, Jesus Del Rincon Martinez, and Abhishek Pramanick (UKFin+ / UKRI, £100k, Nov 2024 – Nov 2025).
  • Anomaly detection on large heterogeneous trading and communication data with Citigroup Belfast (Momentum 1.0, 2022–2023).
  • Tail risk analytics, stress testing and scenario analytics with Funds Axis, Fearghal Kearney, and Colm Kelly (KTP / Innovate UK, £173k, 2021–2023).
  • AI and advanced retail analytics with PearlAI and Byron Graham (KTP / Innovate UK, £165k, 2018–2021).
  • PhD studentship on how AI innovation affects highly valued financial service provision (NI DofE CAST award, £70k, 2021–).
  • Exploring the welfare cost of cultural displays in Northern Ireland using multimodal generative AI, with Declan French (DofE CAST award NI, £70k, 2023–).

Earlier commissioned work (Phoenix Natural Gas demand forecasting, Northern Ireland credit unions landscape review, Invest NI short commissions on predictive analytics and VaR modelling) and other grants listed on the résumé.

Software

Two R packages listed on Zenodo and in the résumé software section: