event description

Policymakers need to assess the current state of the Economy in real-time when only incomplete information is available. Most countries in the COMESA region do not compile high-frequency measures of Gross Domestic Product (GDP) or if they do, the estimates come with a long lag. Yet policy decisions need to be made often, especially when major shocks such as the 2008 GFC, COVID-19, and climate-related shocks have occurred. Publication delays cause missing values at the end of the sample for some of the especially very important variables which complicate forward-looking policy making.

GDP estimates for countries that compile them on a quarterly frequency are released with a considerable lag, while a range of leading and coincident variables are available in a timelier fashion (at a monthly or even higher frequency). Making use of these high-frequency indicators to forecast in real-time the missing values of a higher frequency variable have become a standard in the analysis of economic outlook in what is known as nowcasting.

Nowcasting has become a standard way to bridge the gap between the time lag in data releases and policymaking. Given the growing importance of nowcasting in the overall forecasting and policy analysis systems of central banks and other policy-making institutions, the broad objective of this course is to make officials engaged in policy analysis appreciate nowcasting, acquaint them with the standard nowcasting methods, and introduce machine Learning to nowcasting. 

Target Audience 

The course targets Economists interfacing with Forecasting and Policy Analysis. Participants are expected to have an advanced degree in Economics or relevant experience with monetary policy formulation or implementation.