When emerging markets are addressed, it is easily observable that the widely used models for studying the volatility of stock markets are the ARCH ones. In this particular perimeter, the application of SV models is limited. Therefore, the aim of this paper is to study the volatility of stock prices in several emerging markets from Central and Eastern Europe (CEE), within a stochastic framework. In other words, we use the SV model in order to represent the volatility behaviour of stock market indices from Polish, Croatian, Czech, Romanian and Bulgarian stock exchanges over a five-year period (2012-2017). In addition, we analyse the first ten companies of each index (based on market capitalization) in order to highlight similar and different features of volatility across sectors and industries.
The stochastic volatility model is based on treating the volatility as an unobserved variable, defined as a logarithmic first order autoregressive process. For the construction of the maximum likelihood for the SV model we adopt the Monte Carlo integration method developed by Koopman and Uspensky (2002).
As in the case of previous findings, we expect the estimation results to show a high magnitude of volatility along with a remarkable volatility persistence within the analysed CEE emerging stock exchanges, characterized by a high concentration of stock market activity. In addition, we expect a higher variability within the main sectors: the financial and energy ones, where foreign investors have a large share.