High Dimensional Covariance Matrix Estimation via Bayesian Method

Autor: Si-ming Huang, Jie Tang
Rok vydání: 2014
Předmět:
Zdroj: Advances in Intelligent Systems Research.
ISSN: 1951-6851
DOI: 10.2991/iceeim-14.2014.65
Popis: High-dimensional covariance matrix estimation and its applications in the portfolio selection are increasingly becoming an important topic. However, classical statistical methods used to estimate the sample covariance matrix will lead to inverse covariance matrix biased. Based on this, we propose the high dimensional covariance matrix estimation via Bayesian Method, to ensure that the result of the inverse covariance estimation is unbiased. Index Terms - covariance matrix estimation, portfolios, bayes, unbiasedness
Databáze: OpenAIRE