A0769
Title: From rotational to scalar invariance: Enhancing identifiability in score-driven factor models
Authors: Emilija Dzuverovic - Ca Foscari University of Venice (Italy) [presenting]
Fulvio Corsi - University of Pisa and City University London (Italy)
Giuseppe Buccheri - University of Rome Tor Vergata (Italy)
Abstract: It is shown that, for a certain class of scaling matrices including the inverse square-root of the conditional Fisher information, score-driven factor models are identifiable up to a multiplicative scalar constant under very mild restrictions. This result has no analog in parameter-driven models, as it exploits the different structure of the score-driven factor dynamics. Consequently, score-driven factor models overcome the issue of rotational invariance that typically affects dynamic factor models, thereby enhancing the economic and financial interpretability of the estimated factors. The restrictions are order-invariant and can be generalized to score-driven factor models with dynamic loadings and nonlinear factor models. The identification strategy is tested extensively, using simulated and real data. The empirical analysis on financial and macroeconomic data reveals a substantial increase in log-likelihood ratios and significantly improved out-of-sample forecast performance when switching from the classical restrictions adopted in the literature to the more flexible specifications.
