Are Machine Learning (ML) algorithms superior to traditional econometric models for GDP nowcasting in a time series setting?

Since econometric models have been utilized to generate quantitative forecasts of aggregate economic activity, it is of interest to determine how well the existing models have performed this function.

This paper points to some of the difficulties in large-scale econometric modeling. The major issues here include the logic of the equations, the nature of the stochastic disturbance terms, the logic ...