Home/All Courses/Central Banking/Advanced Econometrics and Time-Series Analysis for Central Bank Economists Training Workshop
Central Banking · Physical, Online
Advanced Econometrics and Time-Series Analysis for Central Bank Economists Training Workshop
10 Days4 sessions available Certificate on completion
This advanced professional workshop equips central bank economists, monetary policy researchers, macroeconomic modellers and quantitative analysts with the econometric techniques required to analyse economic dynamics, develop reliable forecasts and evaluate monetary policy decisions.
The programme focuses on practical applications of advanced time-series analysis, including inflation forecasting, GDP growth modelling, monetary policy transmission, exchange-rate dynamics, interest-rate analysis, macroeconomic shocks and financial stability monitoring. Participants work with modern econometric methods to investigate long-run relationships, estimate dynamic interactions, quantify uncertainty and evaluate how economic variables respond to policy interventions.
The workshop combines economic theory, rigorous statistical testing, practical model estimation and AI-assisted analytical techniques. It is designed to strengthen the quality of empirical research and forecasting used in monetary policy committees, economic research departments and financial stability units.
Course objectives
Apply advanced time-series econometric techniques to analyse inflation, GDP growth, interest rates, exchange rates, credit conditions and other macroeconomic variables.
Translate econometric findings into evidence-based monetary policy recommendations, research papers, forecast reports and policy committee briefings.
Develop robust forecasting and policy simulation models using frequentist, Bayesian and modern AI-assisted approaches.
Investigate short-run dynamics, long-run equilibrium relationships, structural changes and nonlinear responses to economic and monetary policy shocks.
Estimate, diagnose and validate multivariate econometric models for forecasting, monetary policy transmission analysis and economic research.
Who should attend
Central bank economists, monetary policy analysts, macroeconomic researchers, quantitative analysts, econometricians, inflation forecasting specialists, financial stability analysts, financial markets researchers, national accounts statisticians, reserve management analysts, research directors, economic policy advisors and data scientists working in central banks, ministries of finance, regulatory authorities and financial institutions.
Course content
Module 1: Advanced Econometric Foundations for Central Bank Research Review of probability distributions, statistical inference, estimation theory, hypothesis testing, confidence intervals, maximum likelihood estimation and generalised least squares; und
Module 2: Time-Series Properties and Stationarity Stochastic processes, white noise, random walks, deterministic and stochastic trends, stationarity and differencing; autocorrelation and partial autocorrelation functions; unit-root testing using ADF, Phi
Module 3: ARIMA, SARIMA and Advanced Univariate Forecasting Identification, estimation and diagnostic checking of AR, MA, ARMA, ARIMA and SARIMA models; seasonal patterns in inflation and economic activity; information criteria for model selection; rolli
Module 4: Vector Autoregression and Structural VAR Models Specification and estimation of VAR systems, lag selection, stability conditions, impulse response functions and forecast error variance decomposition; structural VAR identification using economic
Module 5: Cointegration and Error-Correction Modelling Long-run equilibrium relationships among economic variables; Engle–Granger and Johansen cointegration tests; vector error-correction models; adjustment coefficients and long-run restrictions; ARDL
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, Dy
Module 7: Bayesian Econometrics and Bayesian VAR Bayesian inference, prior and posterior distributions, likelihood functions, prior selection, Bayesian VAR specification, shrinkage methods, posterior predictive distributions and Bayesian forecast combina
Module 8: State-Space Models and Kalman Filtering State-space representation, latent variables, observation and transition equations, Kalman filtering and smoothing, parameter estimation and time-varying coefficients; estimating unobservable variables su
Module 9: Structural Breaks, Nonlinear Dynamics and Regime Switching Testing for structural breaks using appropriate statistical procedures; multiple-break analysis; Markov-switching and threshold models; time-varying volatility; nonlinear responses to m
Module 10: Volatility Modelling and Financial Time Series ARCH and GARCH models, conditional heteroskedasticity, volatility clustering, asymmetric volatility specifications, financial returns, interest-rate volatility, exchange-rate risk and commodity-pr
Module 11: Local Projections and Causal Analysis of Monetary Policy Local projection methods, impulse response estimation, identification of monetary policy surprises, event-study approaches, instrumental-variable methods where appropriate, counterfactua
Module 12: Mixed-Frequency Econometrics and Dynamic Factor Models Integrating daily, weekly, monthly and quarterly data; MIDAS regressions, bridge equations, dynamic factor models and mixed-frequency state-space approaches; extracting common economic sig
Module 13: Advanced Forecast Evaluation and Model Combination In-sample and out-of-sample evaluation, rolling-origin and expanding-window validation, forecast bias and error measures, directional accuracy, density forecasts and prediction intervals; Dieb
Module 14: Machine Learning, AI and Modern Econometric Workflows Regularised regression, random forests, gradient boosting and selected neural-network approaches for economic time series; feature engineering, lag selection, nonlinear relationships and mo
Module 15: Integrated Central Bank Econometric Research Capstone Participants design and execute an applied econometric study using a realistic macroeconomic dataset. They formulate a research question, prepare the data, test stationarity and cointegrati
General notes
Practical Exercises and Case Studies Unit-root and stationarity testing on inflation and GDP data, ARIMA-based inflation forecasting, VAR estimation using inflation GDP growth and the policy rate, structural VAR analysis of monetary policy shocks, cointegration analysis of money demand and economic activity, estimation of an error-correction model for exchange-rate pass-through, Bayesian VAR forecasting under limited data conditions, Kalman filtering to estimate the output gap, structural-break testing during major economic disruptions, regime-switching analysis of high- and low-inflation periods, GARCH modelling of exchange-rate volatility, local projection analysis of policy-rate transmission, GDP nowcasting using mixed-frequency indicators, comparison of econometric and machine-learning forecasts, and preparation of a central bank research paper with diagnostic tests and policy implications.
Recommended Tools and Technology Stack Econometric software: Stata, EViews, R, MATLAB and Dynare, depending on institutional licensing and participant experience. Python libraries: pandas, NumPy, statsmodels, SciPy, scikit-learn, PyMC and suitable time-series modelling libraries. Bayesian and structural modelling: Bayesian VAR implementations, PyMC, Dynare and relevant institutional modelling frameworks. Data visualisation and reporting: Microsoft Excel, Power BI, Jupyter notebooks, R Markdown and reproducible research reporting tools. Core analytical methods: ARIMA/SARIMA, VAR/SVAR, VECM, ARDL, Bayesian VAR, state-space models, Kalman filtering, GARCH, local projections, dynamic factor models, structural-break tests, regime-switching models and forecast combination. The workshop can be delivered using a single preferred software environment or adapted to the institution’s existing econometric infrastructure.
Practical Outputs and Deliverables Advanced time-series data preparation and diagnostics checklist, econometric model selection framework, stationarity and cointegration testing report, inflation forecasting model, VAR or Bayesian VAR estimation file, monetary policy impulse response analysis, long-run relationship and error-correction model, state-space model for estimating latent economic variables, structural-break and regime-switching analysis, financial volatility model, mixed-frequency GDP nowcasting framework, forecast evaluation and model comparison scorecard, reproducible econometric analysis notebook or script, applied central bank research paper, and policy briefing summarising empirical results, uncertainty and implications for monetary policy.
Research references - Bank for International Settlements (BIS): Economic Modelling and Forecasting Practices in Central Banks. Explains the role of multiple macroeconomic models—including VAR, state-space, Bayesian, semi-structural and DSGE approaches—in policy analysis and forecasting. Bank for International Settlements - BIS, March 2026: BISTRO: A General-Purpose Oracle for Macroeconomic Time Series. Examines emerging AI-based time-series forecasting and its potential use alongside traditional econometric benchmarks. Bank for International Settlements +1 - IMF: Bayesian VARs: A Survey of the Recent Literature. Provides background on Bayesian VAR estimation, forecasting and structural analysis. International Monetary Fund - Journal of Economic Surveys: Central Bank Forecasting: A Survey. Reviews forecasting practices and developments across major central banks.
Frequently asked questions
The workshop is designed for central bank economists, monetary policy researchers, quantitative analysts, macroeconomic modellers and financial stability specialists who use empirical models for economic forecasting and policy analysis.
Participants study ARIMA and SARIMA, VAR and structural VAR, cointegration, error-correction models, Bayesian VAR, state-space models, Kalman filtering, GARCH, local projections, dynamic factor models, structural-break analysis and selected machine-learning approaches.
Yes. Practical applications include inflation forecasting, policy-rate transmission, exchange-rate pass-through, output-gap estimation, money demand, economic shocks, financial volatility and the evaluation of alternative policy scenarios.
A background in statistics, economics or quantitative analysis is recommended. Basic experience with at least one statistical or econometric package will help participants gain the most from the practical sessions. The required programming depth depends on the selected software and course version.
Yes. Participants work through model specification, estimation, diagnostic testing, forecast evaluation and interpretation using practical datasets. The capstone requires participants to complete an applied econometric analysis relevant to central bank research.
The programme examines machine learning and emerging AI forecasting methods as complements to established econometric models. Participants learn to compare predictive performance, identify data leakage, assess model limitations and preserve reproducibility and human oversight.