The next QRFE Webinar will take place on the 25 of March 2021 at 15h00.
Our speaker is Dr. Nazarii Salish from University Carlos III de Madrid. Nazarii's website is here:
Nazarii will present:
Title: Dynamic Factor Model for Functional Time Series: Identification, Estimation, and Prediction
Abstract: A fully functional factor model is proposed in which both the common component and the idiosyncratic component are random elements of the Hilbert space of square-integrable functions on a bounded domain. We discuss in detail the conditions under which the components of such model are exactly and asymptotically identified. This result allows us to obtain a two stage estimation framework for the factors and loading functions as well as separating the common functional component from the idiosyncratic functional component. In particular, we obtain an asymptotically valid criterion that estimates jointly a separation point between common and idiosyncratic components and a dynamic lag structure of the common component. Further, we discuss how to obtain prediction bands under additional distributional assumptions. Finally, the methodology is applied to the problem of yield curve modeling and forecasting. In an out-of-sample experiment, it is shown that predictions can be significantly improved when compared to the predictor from the dynamic Nelson-Siegel model, which is the most commonly used term structure model for yield curves.