2026
How Economics Absorbed AIA field map built from a corpus of 661 sources, read through the formats the discipline argues in — papers, guides, tools, podcasts — with falsifiable predictions about what comes next.
Luka Šikić
I'm an Assistant Professor of Statistics and Data Science at the Croatian Catholic University in Zagreb. My work sits where quantitative methods meet finance, economics, and digital media. I teach statistics, data science, and digital media, and lead research that connects these fields.
Outside the university, I've worked in financial markets and algorithmic trading, and partnered with firms and research institutes on applied work — financial and time-series econometrics, machine learning, NLP-driven media monitoring, and digital analytics. Most of my current attention goes to AI in economic research: how large language models and agentic pipelines change the way empirical work actually gets done, and which parts of the craft they can't replace. Explore How Economics Absorbed AI, my evidence-based retrospective and forecast, or the open seminar and research pipeline behind this work.
A — Now
What I'm working on
- 2026AI in Economics Research — seminar at wiiw, ViennaFrom frontier ideas to a functioning agentic pipeline: slides, a working paper rendered live from Eurostat data, and the full open-source repository.
- 2026DigiKat — seminar at the Complexity Science Hub, ViennaExecutive overview of the project: 710,307 posts across nine platforms, 2021–2026, read through five layers — publisher archetypes, thematic streams, tone and conflict, event spikes, and how influence disperses over time.
- 2026Everything You Want to Know About the State and Public Policies in CroatiaUniversity textbook on public choice and public-sector economics, applied to the Croatian context. With Milan Deskar-Škrbić and Petra Palić. CC BY 4.0.
- 2026Connected but Divided — affective publics on Croatian RedditNetwork, sentiment, and topic analysis of 11,503 Reddit comments around the 2024 elections. With Tamara Kunić. Medijska istraživanja (2026).
- 2025Forecasting returns in thin markets — ML on the Zagreb Stock ExchangeMachine-learning forecasts of weekly returns from 1,100+ features on 2000–2024 data. With Mislav Šagovac and Petra Palić.
B — Selected research
Selected publications
- 2026Connected but Divided: Affective Publics and the Clash of Communities on Croatian Reddit in the Context of the 2024 ElectionsMedijska istraživanja, Vol. 32 No. 1, pp. 31–54
- 2026Competing Media Frames During a National Currency Transition: The Euro Adoption in CroatiaCommunication Today, Vol. 17 No. 1, 2026
- 2025The Breakdown of the Classic Portfolio Hedge: A Markov-Switching Analysis of the US Stock–Bond CorrelationNotitia, Vol. 11 No. 1, pp. 97–111
- 2025Income Convergence of Croatia Within the European Union: An Empirical Analysis of Fractional IntegrationCREBSS, Vol. 11 No. 1, pp. 35–49
C — Background
How I got here
PhD in Economics, University of Zagreb and University of Vienna. MSc in International Economics from the Kiel Institute and MSc in Finance from the University of Zagreb. Before joining the Croatian Catholic University, I spent about eight years as a researcher at the Institute of Social Sciences Ivo Pilar and the Faculty of Croatian Studies. My work appears in international and Croatian journals, and I give invited seminars and hands-on workshops on econometrics, data science, and AI-assisted research.
D — Get in touch
Workshops, consulting, collaborations
I design hands-on workshops in AI for research, applied analytics, and reproducible workflows for universities, institutes, and teams. See previous workshops and public materials, or write to luka.sikic@unicath.hr to discuss a project.