Forecasting in Central Banks

Event Code :MP1B-2020
Venue :Sri Lanka
Host Name :Central Bank of Sri Lanka
Coordinator :Mr. Syaiful Hafizi
Date From :20 Jul 2020
Date To :24 Jul 2020

Descriptions


Forecasting is crucial part of central banks not only from a monetary stability perspective but also from the perspective of financial stability and liquidity policy.  Regardless of the end objective, all central banks/monetary authorities must forecast several macroeconomic, financial and prudential variables to make a timely assessment of the economy and therefore their policy decisions. In this course, we will combine lectures on forecasting theory with hands-on computer-based exercises to forecast variables like inflation, GDP growth and non-performing loans.  External presenters will be Massimiliano Marcellino from Bocconi University in Milan and Romain Lafarguette from the IMF.
 
Objectives
 
At the end of the course, participants will be able to: (1) identify the best forecasting models for macroeconomic, financial and prudential variables; (2) gain experience in forecasting with univariate and multivariate models; (3) be able to perform nowcasts as well as mixed-frequency forecasting; (4) generate fan charts to convey forecast uncertainty; and (5) evaluate forecasting accuracy and other forecast performance measures.
 
Target audience
 
Economists working in the monetary policy, financial stability, liquidity policy and research departments of central banks who are interested in learning about cutting edge statistical and econometric methods that can be used in nowcasting, forecasting, and scenario analysis.  In addition to producing forecasts how to interpret and communicate forecasts will also be covered.
 
 Potential topics
 
•    Econometric modelling of inflation, GDP growth and non-performing loans
•    Forecasting with vector-autoregressions (VARs), Bayesian VARs, conditional forecasting
•    Forecast evaluation and combination, and combination in the presence of asymmetries
•    Mixed-frequency modelling for forecasting
•    Nowcasting and forecasting with large data
•    Density forecasting and evaluations
•    Evaluating long horizon forecasts
•    Unobserved component models with stochastic volatility for forecasting inflation