Dynamic Model Averaging in Large Model Spaces Using Dynamic Occam's Window

TL;DR

This paper introduces a dynamic model averaging method that efficiently handles large model spaces by using a dynamic Occam's window, improving real-time GDP nowcasting in the Euro area.

Contribution

It proposes a novel dynamic Occam's window approach for Bayesian model averaging that scales to large model spaces and adapts over time.

Findings

The method performs well in nowcasting Euro area GDP.

It effectively manages large model spaces without exhaustive enumeration.

Forecasting accuracy compares favorably with existing methods.

Abstract

Bayesian model averaging has become a widely used approach to accounting for uncertainty about the structural form of the model generating the data. When data arrive sequentially and the generating model can change over time, Dynamic Model Averaging (DMA) extends model averaging to deal with this situation. Often in macroeconomics, however, many candidate explanatory variables are available and the number of possible models becomes too large for DMA to be applied in its original form. We propose a new method for this situation which allows us to perform DMA without considering the whole model space, but using a subset of models and dynamically optimizing the choice of models at each point in time. This yields a dynamic form of Occam's window. We evaluate the method in the context of the problem of nowcasting GDP in the Euro area. We find that its forecasting performance compares well that of other methods. Keywords: Bayesian model averaging; Model uncertainty; Nowcasting; Occam's window.