A new model to predict Italian economic growth

Massimiliano Marcellino (Bocconi University and Baffi Centre) and Michael Pfarrhofer (WU University) have just delivered a better model to predict the key performance indicators of Italian economy. The model is described in detail in a discussion paper for Cepr (Center for Economics and Policy Research), whose title is “Nonparametric Mixed Frequency Monitoring Macro-at-Risk”. 

In the paper, Marcellino and Pfarrhofer compare homoskedastic and heteroskedastic mixed frequency (MF) vector autoregression and Bayesian additive regression tree (BART) models to assess their relative performance in predicting tail risk. MF-BART is a nonlinear state space model. They discuss linear approximation approaches to devise computationally efficient estimation algorithms. The models are applied in an out-of-sample backcasting, nowcasting and forecasting exercise for a set of quarterly and monthly macroeconomic variables in Italy: Deficit-to-GDP (Deficit) and Debt-to-GDP (Debt) ratio, real GDP growth (RGDP) as quarterly target variables; unemployment (in differences), industrial production (IP, annualized log differences), and inflation (HICP, annualized log-differences) as monthly targets; Italian long-term interest rates (10-year benchmark), the spread between Italian/German long-term government bond yields, economic sentiment indicator, euro area short-term (3-month maturity) rates, and the USD/EUR exchange rate. 

The dataset comprises quarterly and monthly variables from Italy, ranging from January 2001 until June 2024. The proposed econometric refinements yield improvements in predictive accuracy.