STATISTICS COLLOQUIUM
Sung Hoon Choi
Assistant Professor, Department of Economics
University of Connecticut
Feasible Weighted Projected Principal Component Analysis for Factor Models
with an Application to Bond Risk Premia
Abstract
I develop a feasible weighted projected principal component (FPPC) analysis for factor models in which observable characteristics partially explain the latent factors. This novel method provides more efficient and accurate estimators than existing methods. To increase estimation efficiency, I take into account both cross-sectional dependence and heteroskedasticity by using a consistent estimator of the inverse error covariance matrix as the weight matrix. To improve accuracy, I employ a projection approach using characteristics because it removes noise components in high-dimensional factor analysis. By using the FPPC method, estimators of the factors and loadings have faster rates of convergence than those of the conventional factor analysis. Moreover, I propose an FPPC-based diffusion index forecasting model. The limiting distribution of the parameter estimates and the rate of convergence for forecast errors are obtained. Using U.S. bond market and macroeconomic data, I demonstrate that the proposed model outperforms models based on conventional principal component estimators. I also show that the proposed model performs well among a large group of machine learning techniques in forecasting excess bond returns.
Bio: Dr. Choi’s research focuses on the development of new tools for use with big data, machine learning, and forecasting. The tools primarily involve the development of new theoretical methods for use in both estimation and statistical inference using high-dimensional panel datasets. His research interests include econometric theory, financial econometrics, machine learning, and forecasting with a concentration in high-dimensional data, large panel data and factor models.
Wednesday, October 6, 2021
4:00 p.m. EDT, 1-hour duration
Join by meeting number |
Meeting number (access code): 2620 886 9037 |
Meeting password: HUuFrM922hy |
Join by phone |
+1-415-655-0002 US Toll |
Global call-in numbers
For more information, contact: Tracy Burke at tracy.burke@uconn.edu