A brief survey of density forecasting in macroeconomics

• Main Author: TAY, Anthony S.
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2015
• Subjects: Macroeconomics; economic growth; economic forecasting; density forecasting
• Online Access: https://ink.library.smu.edu.sg/soe_research/1901; http://www.mas.gov.sg/Monetary-Policy-and-Economics/Education-and-Research/Research/Economics-Essays-MR-special-features/2015/A-Brief-Survey-Of-Density-Forecasting-In-Macroeconomics.aspx
• Institution: Singapore Management University

A density forecast of an economic variable is an estimate of the conditional probability density function (p.d.f.) of the possible future values of that variable. For example, a density forecaster might say something like “based on current information, GDP growth over the next year is expected to be normally distributed with mean 3% and standard deviation 0.5%”. A density forecast therefore provides a complete probabilistic description of the possible future realisations of a variable, given some information set. It is a generalisation of the more common point forecast (“GDP growth over the next year is expected to be 3%”) and interval forecast (“GDP growth over the next year is expected to be between 4% and 5%”). This Special Feature presents a brief survey of density forecasting in macroeconomics. It describes density forecasts, and how they can be constructed, presented, evaluated and used. We will begin with a discussion of some of the disadvantages of point and interval forecasts, and some of the benefits of density forecasts. We look at some examples of density forecasts that are already in regular use, and discuss how such forecasts can be evaluated. Finally, we highlight some of the difficulties and challenges of density forecasting.