The role of expectations in modeling and forecasting industrial production

Abstract

This study examines the current need to obtain timely information about the path of macroeconomic aggregates and proposes the modeling and forecasting of industrial activity in Brazil by innovating a production function that takes as arguments, besides proxies for capital and labor, a suitable variable that reflects subjective sentiment about the future business prospects. The modeling strategy for best forecasting relies on VAR models. The results confirm the robustness of the qualitative variables and the entrepreneurs’ expectations, besides showing that the greater the uncertainty over the economic environment is, the more representative will be the role of expectations as a determinant of domestic industrial production.