Εκτίμηση του συστηματικού κινδύνου μετοχών σε ρηχές ευρωπαϊκές αγορές
Estimating the systematic risk of shares in thinner european markets
Σφυρή, Ασπασία Δ.
This dissertation deals with the calculation of betas in thinner European markets. The primary purpose of this study is to examine the predictive ability of betas using different time intervals for calculating periodic stock returns for portfolios of high and low capitalization, respectively. Essentially, the thesis examines the intervalling effect bias in the estimation of the systematic risk of returns of shares traded on the stock market of Greece and France. Specifically, we estimate the systematic risk of stock returns using the market model and applying the OLS method, for the period from 02/01/2002 to 31/12/2012. The results show the presence of the intervalling-effect bias in beta estimates and that the magnitude of this bias is depended on the market value of firms, as well. Furthermore, previous studies related to the market model are presented. In particular, the predictive ability of the model of Hawawini (1983) on the estimation of systemic risk for longer return intervals using shorter return intervals, is examined. Finally, betas are estimated according to the models of Cohen et al (1983a) and Scholes & Williams (1977) and the results of these two models are compared with the results produced running the market model . The results show that there are no statistically significant differences between the mean estimates of the betas with the method of least squares and the models of Cohen et al and Scholes & Williams.