| Abstrakt: |
This research deals with the analysis of different risk measures used for the identification of the rewarded risk factors. Namely, various research has shown that there are some unrewarded risk factors on the market that are relatively easy to expose, which reduces the desired effects of diversification. Because of that, different approaches are used to identify rewarded risk factors. One of them is applied in this research and for that purpose, three different measures of the risk are used. Generally, when the returns distribution is skewed, the downside risk is relevant which could be tested via measure downside beta or coskewness. Downside beta is a modified measure of a systematic risk that takes into account only deviations in stock returns that are below the average market return. Given the fact that this research is conducted on the stocks for which return distribution is mainly positively asymmetric, the third moment is taken into account i.e. skewness. In the context of the contribution of the asymmetry of the individual stock to the asymmetry of a market portfolio, the new measure of risk is taken into account – coskewness. Besides downside beta and coskewness, the third measure used is beta, the standard measure of systematic risk in financial analysis. To determine whether risk measures can be used as a proxy for the expected return and if they have the power to explain cross-sectional variability in stock returns on the Croatian stock market, monthly returns of 64 stocks are used. Stocks included in the analysis were listed on The Zagreb Stock Exchange (ZSE) and included in the CROBEX index at some point in time in the period from January 2005 till April 2019. The applied methodology consists of a time-series analysis of the risk premium and regression analysis: Fama-MacBeth and panel data analysis. The findings show that the downside beta, as a measure of loss aversion did not yield statistically significant risk premiums which raises the question of a loss aversion awareness and reward for it on the Croatian stock market. Also, evidence is found in support of the view that the coskewness could be used as a proxy for the expected return on the Croatian stock market. [ABSTRACT FROM AUTHOR] |
