The Effect of Systematic Risk and Previous Period Returns on Portfolio Selection

There are many techniques to determine investable set of portfolios given return data of assets. However, the theoretical results do not always point out the best portfolios in practice. This is due to the fact that financial dynamics are so difficult to be modelled and this requires many assumptions. The investor may have some preferences to select portfolios. In this study, two selection criteria are proposed to be applied in a mean variance optimization. These criteria are beta coefficient which is a measure of systematic risk and previous period return.The study has an empirical analysis applied on Istanbul Stock Exchange. The findings of the study confirm that these selection criteria may be used to obtain investable portfolios. The analysis with beta selection criteria reveal that the portfolio with lowest 5 beta coefficients is the best alternative. This means that the advantage of low beta which is a natural hedge when stock values declineis superior to diversification benefits of adding new stocks.The previous period return analysis suggest two alternative portfolios. In addition, one of these portfolios generate higher return than the portfolio selected with beta selection criterion. This is also a higher risk portfolio. Therefore the decision is based on risk profile of the investor.This research offer to add selection criteria to the standard approach which is beneficial for academic and practical purposes in portfolio management.