| Autor: |
Aghamohammadi, A., Dadashi, H., Sojoudi, Mahdi, Sojoudi, Meysam, Tavoosi, M. |
| Předmět: |
|
| Zdroj: |
Communications in Statistics: Simulation & Computation; 2024, Vol. 53 Issue 7, p3047-3057, 11p |
| Abstrakt: |
The portfolio optimization problem can be reformulated by a regression model. Mean-variance portfolios, which are constructed using the sample mean and covariance matrix of asset returns can be considered as a function of the ordinary least squares estimator of the linear regression coefficients. It is well known that this estimator has several drawbacks. In particular many researchers have pointed out its high sensitivity in the presence of outliers and its loss of efficiency in the presence of small deviations from the normality assumption. In this paper we propose a novel regression approach for optimizing portfolios by means of quantile regression models. The proposed portfolios are constructed using certain robust and distribution-free estimators and can be obtained by solving a single linear program, where estimation and portfolio optimization are performed in a single step. Our numerical results on simulated and empirical data confirm that the proposed portfolios behave much better than the traditional mean-variance portfolio, in terms of the Sharpe ratio and expected loss in utility. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
