Macroeconomic forecasting during the Great Recession: The return of non-linearity?

Abstract

The debate on the forecasting ability in economics of non-linear models has a long history, and the Great Recession provides us with an opportunity for a re-assessment of the forecasting performance of several classes of non-linear models, widely used in applied macroeconomic research. In this paper, we carry out an extensive analysis over a large quarterly database consisting of major real, nominal and financial variables for a large panel of OECD member countries. It turns out that, on average, non-linear models do not outperform standard linear specifications, even during the Great Recession period. In spite of this result, non-linear models enable to improve forecast accuracy in almost 40% of cases. Especially some countries and/or variables appear to be more adapted to nonlinear forecasting.

The Forecasting Performance of Setar Models: An Empirical Application

“…These gains are particularly evident for the 1‐step‐ahead forecasts, moreover, and in line with the findings by Ferrara et al. (), these advantages are stronger outside the recession period. The rest of the paper is organized as follows.…”

“…Ferrara, Marcellino, and Mogliani () analyse the forecasting performance of nonlinear models (STAR, TAR, time varying specifications and Markov Switching models) for 18 OECD countries and 23 variables. The models are estimated with data from 1970 to 2003 and point forecasts are evaluated over the period 2004q1 to 2009q4, using both fixed moving windows and expanding windows.…”

Does the Great Recession Imply the End of the Great Moderation? International Evidence

“…Among the various explanations of this unexpected severity, Stock and Watson () argue that the macroeconomic shocks were much larger than previously experienced, at least for the United States, especially the shocks associated with financial disruptions and heightened uncertainty. This large shocks hypothesis is also supported in a sense by Ferrara, Marcellino, and Mogliani () who show that the Great Moderation does not come with an increase of the nonlinear dynamics within macroeconomic variables, suggesting thus that a linear behavior with shocks may be more appropriate to describe this specific period of time. Some authors also put forward the major accelerating role of international trade (see Baldwin ), that contributed to the deepness and the worldwide synchronization of this phenomenon…”

Macroeconomic Forecasting in Times of Crises

“…Both the Greenbook and SPF forecasts called for much higher industrial production growth through much of the crisis. Not surprisingly, macroeconomists were criticized for the poor performance of their forecasts (Ferrara, Marcellino, and Mogliani (2015), Del Negro and Schorfheide (2013), Potter (2011)). Given this grim backdrop, in this paper, we develop a forecasting algorithm that improves upon parametric models during rapid changes in the data such as in the 2008/2009 crisis.…”