Econometric Modelling and Nowcasting for Central Banks

Event Code :MP2A-2020
Venue :Philippines
Host Name :Bangko Sentral ng Pilipinas
Coordinator :Ms. Haslina
Date From :24 Aug 2020
Date To :28 Aug 2020


This course focuses on selected modelling and estimation techniques that are increasingly becoming available in econometric software packages and therefore more widely used in central banks.  These approaches include state-space modelling and applications using the Kalman filter; Bayesian estimation of univariate and multivariate models; extensions of VAR analysis to structural approaches such as SVARs and FAVARs; Dynamic factor models. The course also covers volatility estimation from univariate to multivariate approaches, including stochastic volatility. It will also cover selected topics on estimation methods involving panel data. The course combines lectures and hands-on computer-based exercises.
At the end of the course, participants will be able to: (1) specify, estimate and interpret state-space models; (2) estimate, manipulate, and interpret different ways of identifying structural VAR models; (3) specify, estimate and interpret univariate and multivariate models using Bayesian techniques such as the Gibbs sampler; (4) set-up and estimate dynamic factor models; (5) set-up and estimate different volatility models; and (5) handle, estimate and interpret panel-data based methods.
Target Participants
This course is intended for central bank staff whose duty involves significant quantitative analysis and research.  Participants are expected to have excellent quantitative skills and extensive experience using econometric computer programs such as EViews. Some experience in the use of the computer softwares Stata and R is ideal but is not required.  


Potential Topics
  • State space models and Kalman filter
  • Dynamic factor models
  • Univariate and multivariate Bayesian econometrics
  • Structural vector autoregression (SVARs) and extensions
  • Factor-augmented VAR
  • Univariate and multivariate volatility models
  • Panel-data
  • Nowcasting