主講人：蔡恆修 教授 中央研究院統計研究所 副研究員
講    題：Constrained Factor Models
日    期：98年09月29日（星期二）
時    間：下午2:30 –3:20
地    點：科學館S433室
摘    要：
Factor models have been widely used in recent years to improve the accuracy of forecasting
 when many explanatory variables are available. However, the models often encounter the difficulty 
of over-parameterisation and factor interpretation. In this paper, we consider constrained factor analysis 
to obtain a parsimonious factor model and propose likelihood ratio statistics to test the adequacy of factor constraints.
 Real and simulated examples are used to demonstrate the proposed analysis. In an application, we 
show that the constrained factor analysis can provide a better understanding of variations in monthly
 financial asset returns.
(This is a joint work with Ruey Tsay of the University of Chicago)