Short Term Forecasts of Economic Activity: Are Fortnightly Factors Useful?
40 Pages Posted: 12 Jul 2018
Date Written: June 15, 2018
A short term mixed-frequency model is proposed to estimate and forecast the Italian economic activity fortnightly. Building on Frale et al. (2011), we introduce a dynamic factor model with three frequencies (quarterly, monthly and fortnightly), by selecting indicators that show significant coincident and leading properties and are representative of both demand and supply. We find that high-frequency indicators improve the real time forecasts of Italian GDP. Moreover, the model provides a new fortnightly indicator of GDP, consistent with the official quarterly series. Our results emphasize the potential benefit of the high frequency series, providing forecasting gains beyond those based on monthly variables alone.
Keywords: factor models, Kalman filter, temporal disaggregation, mixed frequency data, forecasting
JEL Classification: C53, E17, E32, E37
Suggested Citation: Suggested Citation