Training course on Time Series Analysis and Forecasting for Central Bank Operations

Module 1: Introduction to Time Series in Central Banking

Nature of macroeconomic and financial data

Key applications in policy and operations

Examples: inflation, GDP, interest rates, FX

Module 2: Basics of Time Series Analysis

Stationarity, unit roots, and transformations

Autocorrelation and partial autocorrelation

Model identification and selection criteria (AIC, BIC)

Module 3: Univariate Time Series Models

AR, MA, ARMA processes

ARIMA models and Box-Jenkins methodology

Seasonal models (SARIMA) for inflation and output

Module 4: Multivariate Time Series Models

Vector autoregressions (VAR)

Vector error correction models (VECM)

Applications: policy shocks, monetary transmission

Module 5: Volatility Modeling

ARCH and GARCH models

Extensions: EGARCH, TGARCH, multivariate GARCH

Applications in FX and financial stability

Module 6: State-Space Models and Kalman Filter

Introduction to state-space representation

Kalman filter applications in macroeconomic forecasting

Nowcasting GDP and inflation

Module 7: Forecasting Techniques

Point forecasts vs. density forecasts

Forecast evaluation (RMSE, MAE, Theil’s U)

Forecast combination methods

Module 8: Structural Time Series and Causality

Granger causality testing

Impulse response functions and variance decomposition

Identifying policy shocks

Module 9: High-Frequency and Real-Time Data

High-frequency financial time series

Mixed-frequency models (MIDAS)

Real-time data issues in policy forecasting

Module 10: Time Series Software Applications

Implementing models in EViews, Stata, R, and Python

Automating forecasting routines

Visualization and interpretation of results

Module 11: Case Studies in Central Bank Forecasting

Inflation forecasting and targeting

Exchange rate forecasting

Financial stability stress scenarios

Module 12: Practical Workshop and Capstone Project

Building a forecasting model for inflation or exchange rates

Evaluating model performance and robustnes

Presenting results with policy implications