Macroeconomic Forecasting using Dynamic Factor Models: The Case of Morocco

TL;DR

This paper evaluates the effectiveness of the Factor-Augmented Error Correction Model (FECM) in macroeconomic forecasting for Morocco, demonstrating its superior accuracy and robustness over traditional models using a large dataset.

Contribution

It introduces and empirically tests the FECM model for Moroccan macroeconomic data, highlighting its advantages in capturing economic dynamics and improving forecast accuracy.

Findings

FECM outperforms traditional models in forecasting accuracy.

Inclusion of long-term information improves model performance.

FECM effectively captures economic dynamics in Moroccan data.

Abstract

This article discusses the use of dynamic factor models in macroeconomic forecasting, with a focus on the Factor-Augmented Error Correction Model (FECM). The FECM combines the advantages of cointegration and dynamic factor models, providing a flexible and reliable approach to macroeconomic forecasting, especially for non-stationary variables. We evaluate the forecasting performance of the FECM model on a large dataset of 117 Moroccan economic series with quarterly frequency. Our study shows that FECM outperforms traditional econometric models in terms of forecasting accuracy and robustness. The inclusion of long-term information and common factors in FECM enhances its ability to capture economic dynamics and leads to better forecasting performance than other competing models. Our results suggest that FECM can be a valuable tool for macroeconomic forecasting in Morocco and other similar economies.