Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 161 - Advances in Forecasting of Macroeconomic Variables: New Methods and Applications
Type: Topic Contributed
Date/Time: Tuesday, August 4, 2020 : 10:00 AM to 11:50 AM
Sponsor: Business and Economic Statistics Section
Abstract #312912
Title: Forecasting Real GDP Growth with NIPA Aggregates: A Structural Time Series Approach
Author(s): Kyle Hood* and Baoline Chen and Thomas Triumbur
Companies: Bureau of Economic Analysis and Bureau of Economic Analysis and US Census Bureau
Keywords: model-based filter; business cycles; structural times series; forecasting; granger causality
Abstract:

This paper provides a new set of empirical regularities describing the U.S. macroeconomy, focused on the business cycle fluctuations in real GDP and its major components. The business cycle patterns are assessed using methodology that adapts to the properties of the input series for filtering and avoid creating artificial cycles or other spurious findings that are routine by-products of techniques like the ideal filter. We employ a recent data set from the U.S. national accounts that includes the periods of Great Recession and the ensuing recovery. Using trend and cycle estimates from a model-based filter with desired properties, we exam the strength of the relationship between the aggregate fluctuation and the cyclical components of individual aggregates by examining the cross-autocorrelations to understand whether individual components lead and lag the aggregate cycle; investigating the linkage econometrically via Granger causality tests to see whether cyclical behaviors of individual components are useful in predicting aggregate fluctuations; and evaluating the forecastability of individual components for aggregate real GDP growth.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2020 program