Raul Bâg
Raul Cristian Bâg is a data scientist and academic researcher based at the Humboldt-Universität zu Berlin. His primary areas of expertise lie at the intersection of machine learning, data science, blockchain technology, and Machine Learning Operations (MLOps).
Professional Profile
- Current Role: Academic researcher and data scientist.
- Affiliation: Humboldt-Universität zu Berlin (Faculty of Economics and Business Administration) and the IDA (Institute of Digital Assets).
- Areas of Expertise: Machine learning, MLOps (Machine Learning Operations), blockchain technology, and data architecture
Research and Academic Focus
- Institutional Affiliations: Conducts research at Humboldt-Universität zu Berlin and serves as a researcher at the IDA Institute of Digital Assets affiliated with the Bucharest University of Economic Studies.
- Reproducibility: Develops reproducible data pipelines for scientific research, specifically focusing on Machine Learning Operations für Reproducible Research.
- Data Infrastructure: Investigates decentralized modern data management, including co-authoring core structural work on the Digital Asset Data Lakehouse.
- Financial Risk Analytics: Collaborates on institutional IDA projects such as the Financial Risk Meter (FRM) applied to stock market environments and systemic risk.
Education
- 2021-2026: PhD student in Statistics, Humboldt-Universität zu Berlin
- 2018-2020: M.A. Security & Diplomacy, National School of Political Science and Public Administration, Bucharest
- 2017-2020: B.A. Business Administration, Bucharest Academy of Economic Studies
- 2015-2018: B.A. Political Science, National School of Political Science and Public Administration, Bucharest
Research Papers
- Bag, R., Lessmann, S., Härdle, W. K., & Pele, D. T. (2026). Adapting SHAP to trustworthy window changes. SSRN.
- Zuo, X., Bag, R., & Härdle, W. K. (2025). May the course be with you! Unpublished manuscript.
- Bag, R. C. (2025). Digital asset data lakehouse: The concept based on a blockchain research center. arXiv. https://arxiv.org/abs/2503.15968
- Bag, R., Spilak, B., Winkel, J., & Härdle, W. K. (2025). Quantinar: A blockchain peer-to-peer ecosystem for modern data analytics. Computational Statistics, 40(3), 1361–1396.
- Wojcik, P., Świtała, M., Härdle, W. K., Mare, C., Pele, D. T., Bag, R., Osterrieder, J., et al. (2023). Exploring research visibility of the FinAI COST Action members: A bibliometric analysis of topics. SSRN.
- Poleac, D., & Bag, R. (2023). A design methodology for lifelong career transitions. In Geopolitical perspectives and technological challenges for sustainable development.
- Paraschiv, D. M., Bălășoiu, N., Ben-Amor, S., & Bag, R. C. (2023). Hybridising neurofuzzy model and the seasonal autoregressive model for electricity price forecasting on Germany’s spot market. Amfiteatru Economic, 25(63), 463–478.
- Rădoi, M., Saftiuc, B. P., & Bâg, R. (2023). The impact of new IT&C technologies on academic performance: An analysis of how Web 2.0 and large language models affect the educational and research processes in universities. In Geopolitical perspectives and technological challenges for sustainable development.
- Pele, D. T., Conda, A. I., Bag, R. C., Mazurencu-Marinescu-Pele, M., & Strat, V. A. (2023). Financial risk meter for the Romanian stock market. Romanian Journal of Economic Forecasting, 26(1), 5–20.
- Bag, R. C., Chicu, N., Joga, F. E., & Murafa, C. (2022). The impact of COVID-19 on learning systems–Communication difference and deficiencies in Romanian schools. Research and Science Today, 297–306.
- Căplescu, R. D., Popescu, S. I. A., Bâg, M. R. C., Cristian, M. M. V., & Andrei, M. I. A. Predicția evoluției cazurilor de infecție cu virusul SARS-CoV2 în România.
Open-Source Projects and Courses
Contributes open-source code repositories and advanced analytics modules on the Quantinar platform:
- Statistical Modeling: Instructional content on multivariate analysis, t-tests, proximity measures, and linear regression.
- Geopolitical Sentiment Analysis: Practical data science workflows tracking political rhetoric and shifts through text analysis of major international speeches.
- Social Media Text Mining: Text analysis scripts parsing specific high-profile user patterns, including the most frequent terminology used by tech executives on social platforms.
- Public Health Data Trends: Code repositories that clean and analyze localized social data trends, such as tracking public sentiment during Covid-19 in Switzerland.
- Financial Data Visualizations: Educational assets detailing how to construct interactive dashboards and statistical plots to interpret complex macroeconomic indicators.
…and many more on Quantinar and Quantlet!
Join the Q2 Ecosystem For D.A.D. – Data. Analytics. Dissemination.
