Abstract
In this major paper, we study the influence of structural breaks in the financial market model with high-dimensional data. We present a model which is capable of detecting changes in factor loadings, determining the number of factors and detecting the break date. We consider the case where the break date is both known and unknown and identify the type of instability. For the unknown break date case, we propose a group-LASSO estimator to determine the number of pre- and post-break factors, the break date and the existence of instability of factor loadings when the number of factor is constant. We also present the asymptotic properties of penalized least square estimator with both the cross-sections and the time dimensions tend to infinity.
Further, we develop a cross-validation procedure to obtain the tuning parameters to fine-tune the penalty terms and use the least square approach to estimate the break date after the number of factors is obtained. We also present a Monte Carlo simulation to evaluate the performance of the proposed procedure and analyze real data from 2007-09 of Great Recession. The proposed procedure generally detects the break date correctly during the Great Recession while the procedure performs relatively poorly in estimating the number of factors in the pre- and post-break date case.
Primary Advisor
Dr. Severien Nkurunziza
Program Reader
Dr. Myron Hlynka
Degree Name
Master of Science
Department
Mathematics and Statistics
Document Type
Major Research Paper
Convocation Year
2018