AI-Powered Monetary Policy Analysis, Inflation Forecasting and Nowcasting Workshop/ Seminar
10 Days4 sessions available Certificate on completion
The AI-Powered Monetary Policy Analysis, Inflation Forecasting and Nowcasting Training Course equips central bank professionals, economists, monetary policy analysts and research officers with advanced capabilities in artificial intelligence, machine learning and macroeconomic forecasting. The programme explores how central banks can use AI-powered inflation prediction, real-time GDP nowcasting, high-frequency economic indicators, alternative data and advanced econometric models to strengthen monetary policy analysis. Participants learn to evaluate forecasting models, analyse interest-rate transmission, simulate macroeconomic shocks and develop data-driven decision-support dashboards. With practical applications tailored to central banking and emerging-market economies, the course combines established monetary policy frameworks with modern analytics, explainable AI, scenario modelling and responsible AI governance.
The technical design of this Training reflects established approaches to Forecasting and Policy Analysis Systems (FPAS), alongside emerging central-bank applications of machine learning and generative AI. The IMF has explored machine-learning approaches to inflation forecasting, while the Bank for International Settlements has examined AI-assisted inflation nowcasting using detailed product-price information
Course objectives
1. Develop AI-powered inflation forecasting models using headline and core CPI, food and energy prices, exchange rates, monetary aggregates, inflation expectations and other relevant macroeconomic indicators.
3. Apply advanced econometrics and machine learning to compare ARIMA and VAR models, Bayesian methods, regularised regression, Random Forest, XGBoost and ensemble forecasting approaches.
4. Evaluate monetary policy transmission mechanisms by analysing the effects of policy-rate changes on lending rates, credit growth, exchange rates, aggregate demand, output and inflation.
5. Conduct macroeconomic scenario analysis and policy simulations for supply shocks, currency depreciation, food and fuel price surges, fiscal pressures, external shocks and alternative interest-rate paths.
6. Design an explainable, governed forecasting and decision-support workflow that combines validated models, uncertainty estimates, economic judgement, visual dashboards and policy briefing outputs for monetary policy committees.
2. Build real-time GDP nowcasting systems that combine mixed-frequency economic indicators, dynamic factor models, machine learning and alternative data to assess economic activity before official statistics are released.
Who should attend
Central bank monetary policy officers, monetary policy analysts, central bank economists, macroeconomic research officers, inflation analysis specialists, inflation forecasting teams, economic modelling specialists, GDP forecasting and nowcasting analysts, national accounts statisticians, economic statistics professionals, quantitative economists, econometricians, data scientists, machine learning engineers, AI specialists in financial institutions, monetary policy committee secretariat staff, financial markets analysts, monetary operations officers, interest rate transmission analysts, macro-financial surveillance specialists, financial stability analysts, economic research managers, senior central bank executives, policy advisers, ministry of finance economists, national treasury officials, national statistics office professionals, monetary authorities and financial sector regulators.
Course content
Module 1. Central Bank Monetary Policy Frameworks and FPAS Monetary policy objectives and transmission channels, inflation-targeting frameworks, monetary aggregates and interest-rate-based policy regimes, macroeconomic stability, output gaps, potential GD
Module 2. Inflation Dynamics and CPI Analytics Headline and core inflation, CPI basket weights, monthly and annual inflation rates, seasonality and calendar effects, trimmed-mean and weighted-median inflation, inflation persistence, disaggregated food and
Module 3. AI-Powered Inflation Forecasting Forecasting headline and core CPI, inflation prediction at one-, three-, six- and twelve-month horizons, autoregressive benchmarks, LASSO and Elastic Net, Random Forest, XGBoost, nonlinear relationships, ensemble
Module 4. Real-Time GDP Nowcasting Quarterly GDP estimation before official publication, mixed-frequency data integration, monthly industrial production, trade, electricity consumption, tax receipts and business surveys, Dynamic Factor Models, bridge equa
Module 5. High-Frequency Indicators and Alternative Data Electronic payments and transaction aggregates, mobile money indicators where legally accessible, fuel and electricity consumption, customs and shipping data, satellite-derived night-time lights, po
Module 6. Advanced Econometrics and Time-Series Forecasting Stationarity and differencing, autocorrelation, ARIMA and SARIMA, VAR and structural VAR, cointegration and error-correction models, state-space models, Kalman filtering, Bayesian estimation, Dyn
Module 7. Machine Learning for Macroeconomic Forecasting Supervised learning for continuous economic variables, lagged features, rolling-window features, regularisation, Random Forest, gradient boosting, XGBoost, LightGBM, model ensembles, hyperparameter
Module 9. Exchange Rate Pass-Through and Imported Inflation Exchange-rate depreciation and appreciation, imported inflation, global oil and food prices, freight and supply-chain costs, currency volatility, exchange-rate pass-through estimation, commodity
Module 10. Macroeconomic Scenario Analysis and Policy Simulation Baseline and alternative projections, adverse and upside scenarios, drought and food-supply shocks, oil-price increases, currency depreciation, fiscal expansion, external demand shocks, inte
Module 11. Inflation Expectations and AI-Powered Text Analytics Household and business inflation expectations, financial market indicators, surveys, central bank statements, economic news, sentiment classification, natural language processing, topic model
Module 12. Explainable AI and Model Risk Management Rolling-origin backtesting, RMSE and MAE, benchmark comparisons, forecast bias, probabilistic forecasts, prediction intervals, calibration, SHAP-based explanations where appropriate, model stability, dri
Module 13. Economic Intelligence Dashboards and Automated Forecasting Python-based analytical workflows, reproducible data preparation, scheduled model updates, forecasting APIs, Power BI and Plotly dashboards, inflation heatmaps, GDP nowcast tracking, fo
Module 14. AI Governance, Data Security and Responsible AI Confidential economic data management, access controls, secure model development, data lineage, audit trails, reproducible research, third-party AI risk, privacy-preserving analytics, bias assessm
Module 15. Integrated Monetary Policy Forecasting Capstone Develop an end-to-end inflation forecast, produce a GDP nowcast, compare econometric and machine-learning models, assess interest-rate transmission, simulate adverse macroeconomic shocks, generate
General notes
Authoritative references: These resources provide a sound foundation for updating the course as central bank practices and AI methods evolve. - International Monetary Fund — Model-Based Monetary Policy Analysis and Forecasting. Reference for FPAS, semi-structural models and policy-consistent forecasting. - Bank for International Settlements — Project Spectrum: Using Generative AI to Enhance Inflation Nowcasting. Relevant to product-level price data and AI-assisted inflation nowcasting. - IMF — Mending the Crystal Ball: Enhanced Inflation Forecasts with Machine Learning. Research on ML forecasting methods and comparisons with traditional benchmarks. - IMF — Deus ex Machina? A Framework for Macro Forecasting with Machine Learning. A framework for machine-learning-based macroeconomic forecasting and model ensembles. - statsmodels documentation and scikit-learn documentation. Technical references for econometric and machine-learning laboratories.
AI-powered monetary policy analysis, inflation forecasting training, GDP nowcasting course, central bank AI training, machine learning for economists, monetary policy simulation, macroeconomic forecasting, central bank data analytics, inflation prediction models, AI for monetary authorities
Expected participant outputs By the end of the course, each participant or working group should aim to produce: - An inflation forecasting model with benchmark comparisons and documented limitations. - A GDP nowcasting prototype using mixed-frequency economic indicators. - A model evaluation report with out-of-sample errors and uncertainty estimates. - A monetary policy transmission analysis or scenario simulation. - An interactive macroeconomic monitoring dashboard. - A draft AI model governance and validation checklist. - A policy briefing summarising the baseline outlook, key risks and alternative scenarios.
AI-powered inflation forecasting, real-time GDP nowcasting, macroeconomic data preparation and analysis, econometric and machine learning model development, CPI and commodity price analysis, exchange rate and interest rate transmission analysis, high-frequency and alternative data analytics, inflation expectations and sentiment analysis, macroeconomic scenario modelling and policy simulations, forecast validation and uncertainty analysis, explainable AI and model governance, automated forecasting pipelines, central bank dashboards and policy briefings, macroeconomic risk monitoring and integrated forecasting capstone project.
Recommended prerequisites: Familiarity with macroeconomics, monetary policy and basic statistics. Prior experience with Python, R, MATLAB, EViews or comparable analytical software is helpful for the extended technical version. A non-coding executive version can focus on interpreting forecasts, scenario outputs, model governance and policy decisions.
Frequently asked questions
It is the application of machine learning, econometrics and data analytics to help central banks interpret inflation dynamics, forecast economic activity, assess monetary policy transmission and evaluate alternative economic scenarios.
Nowcasting estimates the current or very near-term state of inflation using the latest available information, including indicators that may be released before official CPI statistics. Forecasting estimates future inflation over specified horizons.
The technical programme uses Python, pandas, statsmodels, scikit-learn, XGBoost, Bayesian modelling tools and dashboard technologies. The institutional version can also incorporate MATLAB/Octave and the IRIS Toolbox where relevant to existing macroeconomic modelling workflows.
Yes. Case studies can address food-price volatility, exchange-rate pass-through, fuel imports, weather-related supply shocks, mobile-money indicators, informal economic activity and data-release constraints. The final curriculum should reflect each institution's monetary policy framework and available data.
No. It teaches participants to develop and validate forecasting models and decision-support tools. Policy decisions remain the responsibility of authorised officials and committees, supported by economic judgement, institutional mandates and transparent assessment of uncertainty.
It combines monetary economics with hands-on AI forecasting, alternative data, mixed-frequency nowcasting, explainable machine learning, policy simulation, model validation and deployment-oriented workflows.