### Abstract

The portfolio beta β_{p} is quite an important coefficient in modern portfolio theory since it efficiently measures portfolio volatility relative to the benchmark index or the capital market. β_{p} is usually employed for portfolio evaluation or prediction but scarcely for portfolio construction process. The main objective of this paper is to propose a portfolio algorithm that engages β_{p} in its portfolio construction process and studies its strengths. Our portfolio algorithm termed as β-G portfolio algorithm selects stocks based on their market capitalization and optimizes them in terms of the standard deviation of β_{p}. The optimizing process or finding optimal weights is done by the genetic algorithm. Our major findings on β-G portfolio algorithm are: (i) its performance depends on market volatility, i.e. it is expected to work well for a stable market whether it is bullish or bearish (ii) it tends to register outstanding performance for short-term applications.

Original language | English |
---|---|

Pages (from-to) | 527-534 |

Number of pages | 8 |

Journal | Expert Systems with Applications |

Volume | 30 |

Issue number | 3 |

DOIs | |

Publication status | Published - 2006 Apr 1 |

### Fingerprint

### All Science Journal Classification (ASJC) codes

- Engineering(all)
- Computer Science Applications
- Artificial Intelligence

### Cite this

*Expert Systems with Applications*,

*30*(3), 527-534. https://doi.org/10.1016/j.eswa.2005.10.010

}

*Expert Systems with Applications*, vol. 30, no. 3, pp. 527-534. https://doi.org/10.1016/j.eswa.2005.10.010

**Portfolio algorithm based on portfolio beta using genetic algorithm.** / Oh, Kyong Joo; Kim, Tae Yoon; Min, Sung Hwan; Lee, Hyoung Yong.

Research output: Contribution to journal › Article

TY - JOUR

T1 - Portfolio algorithm based on portfolio beta using genetic algorithm

AU - Oh, Kyong Joo

AU - Kim, Tae Yoon

AU - Min, Sung Hwan

AU - Lee, Hyoung Yong

PY - 2006/4/1

Y1 - 2006/4/1

N2 - The portfolio beta βp is quite an important coefficient in modern portfolio theory since it efficiently measures portfolio volatility relative to the benchmark index or the capital market. βp is usually employed for portfolio evaluation or prediction but scarcely for portfolio construction process. The main objective of this paper is to propose a portfolio algorithm that engages βp in its portfolio construction process and studies its strengths. Our portfolio algorithm termed as β-G portfolio algorithm selects stocks based on their market capitalization and optimizes them in terms of the standard deviation of βp. The optimizing process or finding optimal weights is done by the genetic algorithm. Our major findings on β-G portfolio algorithm are: (i) its performance depends on market volatility, i.e. it is expected to work well for a stable market whether it is bullish or bearish (ii) it tends to register outstanding performance for short-term applications.

AB - The portfolio beta βp is quite an important coefficient in modern portfolio theory since it efficiently measures portfolio volatility relative to the benchmark index or the capital market. βp is usually employed for portfolio evaluation or prediction but scarcely for portfolio construction process. The main objective of this paper is to propose a portfolio algorithm that engages βp in its portfolio construction process and studies its strengths. Our portfolio algorithm termed as β-G portfolio algorithm selects stocks based on their market capitalization and optimizes them in terms of the standard deviation of βp. The optimizing process or finding optimal weights is done by the genetic algorithm. Our major findings on β-G portfolio algorithm are: (i) its performance depends on market volatility, i.e. it is expected to work well for a stable market whether it is bullish or bearish (ii) it tends to register outstanding performance for short-term applications.

UR - http://www.scopus.com/inward/record.url?scp=30944453897&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=30944453897&partnerID=8YFLogxK

U2 - 10.1016/j.eswa.2005.10.010

DO - 10.1016/j.eswa.2005.10.010

M3 - Article

AN - SCOPUS:30944453897

VL - 30

SP - 527

EP - 534

JO - Expert Systems with Applications

JF - Expert Systems with Applications

SN - 0957-4174

IS - 3

ER -