Linlin Niu visited the National University of Singapore from August 28th to September 7th 2018. Niu and her co-authors are team members of the international research project "Quantitative Easing and Financial (In)Stability" funded by Volkswagen Foundation.
Niu used her research stay to work on the resubmission of the paper "Sparse-Group Independent Component Analysis with Application to Yield Curve Prediction" co-authored with Qiang He and Ying Chen (National University of Singapore) and Ray-Bing Chen (National Cheng Kung University), which was accepted for publication at the Computational Statistics and Data Analysis (Vol. 133, May 2019). During her research stay, Niu discussed her further collaboration on adaptive forecast method on yield curve prediction under an unconventional policy.
In their paper "Sparse-Group Independent Component Analysis with Application to Yield Curve Prediction", the authors propose a Sparse-Group Independent Component Analysis (SG-ICA) method to extract independent factors from high dimensional multivariate data. The method provides a unified and flexible framework that automatically identifies the number of factors and simultaneously estimates a sparse loading matrix, enabling them to discover important features and offers improved interpretability of the estimators. They establish the consistency and asymptotic normality of the loading matrix estimator, demonstrate its finite sample performance with simulation studies, and illustrate its application using the daily US Overnight Index Swap rates from Oct 2011 to Mar 2015 with 15 maturities ranging from 1 week to 30 years. With higher efficiency of extracting factors, the forecasting performance of the SG-ICA is remarkably better than the popular parametric DNS model in an era of quantitative easing with the short-term interest rate being close to zero.