Abstract
A novel class of dimension reduction methods is combined with a stochastic multi-factor panel regression-based state-space model in order to model the dynamics of yield curves whilst incorporating regression factors. This is achieved via Probabilistic Principal Component Analysis (PPCA) in which new statistically-robust variants are derived also treating missing data. We embed the rank reduced feature extractions into a stochastic representation for state-space models for yield curve dynamics and compare the results to classical multi-factor dynamic Nelson–Siegel state-space models. This leads to important new representations of yield curve models that can be practically important for addressing questions of financial stress testing and monetary policy interventions, which can incorporate efficiently financial big data. We illustrate our results on various financial and macroeconomic datasets from the Euro Zone and international market.
Highlights
The increased collection and accessibility to complex financial and macroeconomic data are transforming the way in which financial services operate
In order to handle the presence of irregularities in the real data time series and the possibility of outliers, we develop a novel statistically-robust class of methods for feature extraction based on probabilistic principal component analysis introduced by Tipping and Bishop (1999)
We choose to focus on robust alternatives to the Gaussian Principal Component Analysis (PPCA), and we demonstrate how to use the t-Student distribution in order to account for heavy tail assumptions, which result in the methodology being statistically robust to the distant observations in a sample set of outliers
Summary
The increased collection and accessibility to complex financial and macroeconomic data are transforming the way in which financial services operate. The increasing volume of market data poses both an opportunity and a challenge for financial institutions to deepen their understanding of market-wide and country-specific sources of risk present in financial markets and their implications for modelling and forecasting markets’ dynamics. Such information is of key importance for efficient investment decision making, understanding the influence of unconventional monetary policies and risk management procedures such as stress testing. We focus on providing a coherent methodology that utilises global macroeconomic and financial market datasets in modelling one of the most popular indicators of economic activity, the London Interbank Offered Rate (Libor). Econometrics 2018, 6, 34 datasets in a meaningful and parsimonious way to help explain the dynamics of Libor, we apply the feature extraction methodology with special tailoring for handling irregular time series (missing data) and outliers
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
