Big Data and Machine Learning in Banking & Central Banking Training Workshop

Course overview

The Big Data and Machine Learning in Central Banking Training Workshop equips central bank professionals with practical skills to leverage large-scale datasets, advanced analytics and machine learning for monetary policy analysis, economic forecasting, financial stability surveillance, banking supervision and evidence-based policy development.

The programme explores how central banks can combine traditional macroeconomic and financial statistics with high-frequency payments data, electronic price information, banking transactions, credit registry records, financial market data, business surveys, satellite imagery and text-based information to improve economic intelligence and identify emerging risks. Participants learn how to establish analytical workflows that transform complex data into actionable insights for policy committees, research departments, supervisory teams and financial stability authorities.

Course objectives

• Explain big data architecture, data engineering and machine learning applications relevant to central banking, monetary policy and financial regulation.
• Design responsible, secure and explainable analytical workflows that support policy research, financial stability monitoring and data-driven decision-making.
• Apply AI-driven anomaly detection, natural language processing and predictive analytics to identify emerging economic and financial vulnerabilities.
• Develop and evaluate machine-learning models for economic nowcasting, financial risk detection, payment analytics and supervisory surveillance.
• Identify, acquire, integrate and prepare structured and unstructured data from economic, financial, transactional, market and alternative sources.

Who should attend

Course content

• Module 1: Big Data and Machine Learning Foundations for Central Banking The big data landscape in central banking, characteristics of high-volume and high-velocity data, structured and unstructured information, the five Vs of big data, traditional statis
• Module 2: Central Bank Data Ecosystems and Big Data Architecture Central bank data landscapes, enterprise data architecture, operational and analytical databases, data lakes, data warehouses and lakehouses, batch and streaming data processing, data inges
• Module 3: Data Engineering, ETL/ELT and Large-Scale Data Processing Data extraction, transformation and loading, ETL and ELT architecture, ingestion from regulatory returns and financial databases, data cleansing and validation, schema management, duplic
• Module 4: Alternative Data and High-Frequency Economic Intelligence High-frequency data for economic monitoring, electronic payment aggregates, point-of-sale transactions, mobile money indicators, card spending data, electronic price information, web-scr
• Module 5: Data Governance, Quality, Privacy and Cybersecurity Central bank data governance frameworks, data ownership and stewardship, data classification, metadata and data dictionaries, data quality dimensions, accuracy and completeness checks, data li
• Module 6: Machine Learning Methods for Economic and Financial Analysis Regression and classification algorithms, linear and logistic regression, decision trees, random forests, gradient boosting, XGBoost and LightGBM, support vector machines, k-means clu
• Module 7: Machine Learning for Inflation Forecasting and Economic Nowcasting Macroeconomic data preparation, inflation and core inflation forecasting, GDP nowcasting, mixed-frequency data integration, dynamic factor models, bridge equations, regularised
• Module 8: AI for Monetary Policy Analysis and Economic Research Machine learning for monetary policy research, policy rate and inflation relationships, exchange-rate pass-through, interest-rate transmission, credit growth and economic activity, financial
• Module 9: Big Data and Machine Learning for Banking Supervision Regulatory reporting analytics, large-scale banking data integration, bank-level and transaction-level risk indicators, credit portfolio deterioration, non-performing loan prediction, liquid
• Module 10: AI-Powered Payment Systems Analytics and Financial Crime Detection Retail and wholesale payment system data, payment system performance analytics, transaction network analysis, unusual transaction detection, fraud pattern recognition, payment
• Module 11: Financial Stability, Systemic Risk and Network Analytics Financial interconnectedness, interbank exposures, financial network structures, counterparty concentration, network centrality measures, contagion channels, payment system dependencies,
• Module 12: Natural Language Processing and Generative AI for Central Banks Natural language processing foundations, text preprocessing, tokenisation, embeddings, topic modelling, sentiment analysis, named entity recognition, document classification, info
• Module 13: Big Data Platforms, Cloud Analytics and Scalable AI Deployment Enterprise big data platforms, cloud and hybrid analytical environments, data lakehouse architecture, distributed data processing, batch and streaming analytics, Apache Spark, Kafk
• Module 14: Explainable AI, Model Risk Management and Responsible Deployment Model lifecycle management, model documentation, training data provenance, model validation, backtesting, performance drift, concept drift, explainable AI, feature importance, SH
• Module 15: Capstone — Developing a Big Data and Machine Learning Solution for a Central Bank Identification of a central banking problem, definition of the analytical use case, data inventory and access assessment, preparation of a secure analytical da

General notes

• The workshop covers big data architecture, data engineering, data quality, machine learning, anomaly detection, natural language processing, forecasting, network analytics, explainable AI and model governance. Practical applications include inflation nowcasting, GDP forecasting, payment system surveillance, credit risk analysis, financial fraud detection, systemic risk monitoring and the analysis of central bank communications.
• Special attention is given to the realities of central banks in emerging and developing economies, including fragmented datasets, limited historical observations, data confidentiality, financial inclusion, digital payment expansion, changing economic structures and the need for robust, explainable and reproducible analytical models.
• The course integrates technical practice with central bank policy requirements, drawing on relevant research from the Bank for International Settlements (BIS), the International Monetary Fund (IMF) and other institutions. It emphasises that big data and machine learning complement, rather than automatically replace, established economic theory, econometric analysis and expert policy judgement
• Practical Exercises and Case Studies Build a high-frequency economic activity indicator using sample payment, price or business survey data. Prepare a data pipeline that integrates regulatory reporting data with macroeconomic indicators. Develop an inflation forecasting model and compare machine learning with an econometric benchmark. Create a GDP nowcasting framework using mixed-frequency indicators. Use anomaly detection to identify unusual movements in simulated banking returns. Analyse a synthetic payment transaction dataset to identify abnormal transaction patterns. Build a bank risk classification model using financial ratios and selected supervisory indicators. Conduct network analysis on simulated interbank exposures to identify concentration and contagion risks. Apply natural language processing to central bank statements or economic reports. Develop an explainable AI report showing the main factors behind model predictions. Design a financial stability or supervisory monitoring dashboard. Present a capstone solution with data governance, model validation and implementation recommendations.
• Recommended Software, Tools and Platforms Big data and data engineering: Apache Spark, Apache Kafka, Apache Flink, SQL, Python, Azure Data Factory, Informatica, IBM DataStage.
• Enterprise data platforms: Microsoft Fabric, Databricks, Snowflake, Amazon Web Services, Microsoft Azure, Google Cloud Platform.
• Machine learning and AI: Python, pandas, NumPy, scikit-learn, XGBoost, LightGBM, TensorFlow, PyTorch, MLflow.
• Econometrics and statistical research: Stata, EViews, R, SAS, Python statsmodels.
• NLP and generative AI: Hugging Face Transformers, spaCy, approved enterprise AI assistants, vector databases and retrieval-augmented generation frameworks.
• Data governance and security: Microsoft Purview, Collibra, enterprise identity and access management, data catalogues, encryption and audit logging tools.
• Dashboards and visualisation: Microsoft Power BI, Tableau, Plotly, Matplotlib, Grafana.
• Research and data sources: Approved central bank databases, regulatory returns, payment system aggregates, credit registry data, official macroeconomic statistics, market data and appropriate alternative data providers.
• The final technology stack should reflect institutional architecture, licensing, security requirements, data residency, analytical maturity and the availability of suitable training datasets. Practical exercises can be delivered using synthetic or anonymised data.
• Expected Practical Outputs Central bank big data use-case inventory, data source catalogue, data dictionary, data quality assessment framework, secure data integration plan, ETL/ELT workflow design, macroeconomic indicator dataset, high-frequency economic activity indicator, inflation forecasting model, GDP nowcasting framework, machine-learning benchmark report, bank risk scoring prototype, regulatory data anomaly detection framework, payment analytics report, financial transaction anomaly detection prototype, financial network analysis, systemic risk indicator dashboard, central bank text analytics prototype, AI-assisted research workflow, model validation checklist, explainable AI report, model governance documentation, data privacy and security checklist, analytical dashboard specification, capstone technical report and institutional implementation roadmap.
• Big Data Analytics for Central Banks, Machine Learning for Central Banking, AI for Central Banks, Big Data in Monetary Policy, Machine Learning for Banking Supervision, AI for Financial Stability, Economic Nowcasting with Machine Learning, Central Bank Data Analytics Training, AI for Payment Systems, Financial Risk Analytics Training, Big Data for Monetary Policy Research, Machine Learning for Financial Regulation.

Frequently asked questions

Schedule & fees