DocumentCode :
3256977
Title :
Factor model estimation by using the alpha-EM algorithm
Author :
Tengjie Jia
Author_Institution :
Quantitative Finance, Appl. Math & Stat, Stony Brook Univ., Stony Brook, NY, USA
fYear :
2013
fDate :
3-5 Dec. 2013
Firstpage :
1143
Lastpage :
1143
Abstract :
In this paper, we present the alpha-EM algorithm for factor model estimation for given sample covariance. The alpha-EM includes the traditional log-EM as its proper subset. Since we use log-EM for factor analysis, however, it is shown that alpha-EM can also be used on factor analysis and more important the convergence speed of the alpha-EM is much faster than log-EM. It also allows us to choose different alphas to achieve the fastest convergence speed and more accurate factor model estimation.
Keywords :
convergence; covariance analysis; expectation-maximisation algorithm; alpha-EM algorithm; factor analysis; factor model estimation; fastest convergence speed; log-EM; sample covariance; Algorithm design and analysis; Computational modeling; Convergence; Equations; Estimation; Load modeling; Mathematical model;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Global Conference on Signal and Information Processing (GlobalSIP), 2013 IEEE
Conference_Location :
Austin, TX
Type :
conf
DOI :
10.1109/GlobalSIP.2013.6737106
Filename :
6737106
Link To Document :
بازگشت