Portfolio selection based on semivariance and distance correlation under minimum variance framework

Ruili Sun, Tiefeng Ma, Shuangzhe Liu

Research output: Contribution to journalArticle

Abstract

In the minimum variance model, the covariance matrix plays an important role because it measures the risk and relationship of asset returns simultaneously under the normality assumption. However, in practice, the distribution of asset returns is nonnormal and has an obvious fat-tail nature. In addition, the risk is one-sided. In this paper, the main objective is to propose a better tool to replace the covariance matrix. The covariance matrix can be decomposed into two parts: a diagonal variance matrix and a square matrix with its elements being the Pearson correlation coefficient. A substitution of the covariance matrix is presented by replacing the variance and Pearson correlation coefficient in the decomposition of the covariance matrix with a semivariance and distance correlation coefficient, respectively. The proposed portfolio optimization strategy is applied to empirical data, and the numerical studies show the strategy performs well.

Original languageEnglish
Pages (from-to)373-394
Number of pages22
JournalStatistica Neerlandica
Volume73
Issue number3
Early online date21 Mar 2019
DOIs
Publication statusPublished - Aug 2019

Fingerprint

Minimum Variance
Portfolio Selection
Covariance matrix
Correlation coefficient
Pearson Correlation
Fat Tails
Portfolio Optimization
Square matrix
Normality
Substitution
Numerical Study
Framework
Semivariance
Minimum variance
Portfolio selection
Decompose

Cite this

@article{2108242daf9d46b7aae0abf33ef85d81,
title = "Portfolio selection based on semivariance and distance correlation under minimum variance framework",
abstract = "In the minimum variance model, the covariance matrix plays an important role because it measures the risk and relationship of asset returns simultaneously under the normality assumption. However, in practice, the distribution of asset returns is nonnormal and has an obvious fat-tail nature. In addition, the risk is one-sided. In this paper, the main objective is to propose a better tool to replace the covariance matrix. The covariance matrix can be decomposed into two parts: a diagonal variance matrix and a square matrix with its elements being the Pearson correlation coefficient. A substitution of the covariance matrix is presented by replacing the variance and Pearson correlation coefficient in the decomposition of the covariance matrix with a semivariance and distance correlation coefficient, respectively. The proposed portfolio optimization strategy is applied to empirical data, and the numerical studies show the strategy performs well.",
keywords = "distance correlation, downside risk, fat-tail, portfolio optimization, semivariance",
author = "Ruili Sun and Tiefeng Ma and Shuangzhe Liu",
year = "2019",
month = "8",
doi = "10.1111/stan.12174",
language = "English",
volume = "73",
pages = "373--394",
journal = "Statistica Neerlandica",
issn = "0039-0402",
publisher = "Wiley-Blackwell",
number = "3",

}

Portfolio selection based on semivariance and distance correlation under minimum variance framework. / Sun, Ruili; Ma, Tiefeng; Liu, Shuangzhe.

In: Statistica Neerlandica, Vol. 73, No. 3, 08.2019, p. 373-394.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Portfolio selection based on semivariance and distance correlation under minimum variance framework

AU - Sun, Ruili

AU - Ma, Tiefeng

AU - Liu, Shuangzhe

PY - 2019/8

Y1 - 2019/8

N2 - In the minimum variance model, the covariance matrix plays an important role because it measures the risk and relationship of asset returns simultaneously under the normality assumption. However, in practice, the distribution of asset returns is nonnormal and has an obvious fat-tail nature. In addition, the risk is one-sided. In this paper, the main objective is to propose a better tool to replace the covariance matrix. The covariance matrix can be decomposed into two parts: a diagonal variance matrix and a square matrix with its elements being the Pearson correlation coefficient. A substitution of the covariance matrix is presented by replacing the variance and Pearson correlation coefficient in the decomposition of the covariance matrix with a semivariance and distance correlation coefficient, respectively. The proposed portfolio optimization strategy is applied to empirical data, and the numerical studies show the strategy performs well.

AB - In the minimum variance model, the covariance matrix plays an important role because it measures the risk and relationship of asset returns simultaneously under the normality assumption. However, in practice, the distribution of asset returns is nonnormal and has an obvious fat-tail nature. In addition, the risk is one-sided. In this paper, the main objective is to propose a better tool to replace the covariance matrix. The covariance matrix can be decomposed into two parts: a diagonal variance matrix and a square matrix with its elements being the Pearson correlation coefficient. A substitution of the covariance matrix is presented by replacing the variance and Pearson correlation coefficient in the decomposition of the covariance matrix with a semivariance and distance correlation coefficient, respectively. The proposed portfolio optimization strategy is applied to empirical data, and the numerical studies show the strategy performs well.

KW - distance correlation

KW - downside risk

KW - fat-tail

KW - portfolio optimization

KW - semivariance

UR - http://www.scopus.com/inward/record.url?scp=85068257618&partnerID=8YFLogxK

U2 - 10.1111/stan.12174

DO - 10.1111/stan.12174

M3 - Article

VL - 73

SP - 373

EP - 394

JO - Statistica Neerlandica

JF - Statistica Neerlandica

SN - 0039-0402

IS - 3

ER -