تغییرات ناگهانی در نوسانات: بررسی موردی پنج بازار سهام در اروپای مرکزی

This paper investigates sudden changes in volatility in the stock markets of new European Union (EU) members by utilizing the iterated cumulative sums of squares (ICSS) algorithm. Using weekly data over the sample period 1994–2006, the time period of sudden change in variance of returns and the length of this variance shift are detected. A sudden change in volatility seems to arise from the evolution of emerging stock markets, exchange rate policy changes and financial crises. Evidence also reveals that when sudden shifts are taken into account in the GARCH models, the persistence of volatility is reduced significantly in every series. It suggests that many previous studies may have overestimated the degree of volatility persistence existing in financial time series.

In this paper we examine sudden changes of volatility in the stock markets of new EU members, which were experienced during the transition period from command to market economy and during the period of integration into the EU. Theory suggests that structural changes in fundamentals are associated with the changes in the behaviour of stock markets, since stock prices theoretically reflect expectations of future dividends, interest rates and risk premia, which in turn depend on macroeconomic conditions. It follows that both first and second moments of stock returns should be affected during the process of transition and integration of these economies to the extent that it affects fundamentals (Morana and Beltratti, 2002). The transition of economic systems followed by the currently on-going process of economic integration within the EU may have considerably affected the fundamentals in the new EU member states. Given the existing empirical literature on the shift in the stock market on such occasions,1 it is worthwhile addressing the issue of the pattern of volatility in stock returns for the new EU member states. Time varying volatility of stock returns has been extensively modelled by the GARCH with high frequency stock data to find high persistence in volatility. The GARCH approach assumes that there is no shift in volatility; however, in such emerging markets there may potentially be sudden shifts in volatility. It is therefore important to take account of these shifts in estimating volatility persistence. In this paper, the shifts in volatility are identified by utilizing the iterated cumulative sums of squares (ICSS) algorithm of Inclan and Tiao (1994). The GARCH model is then estimated by taking account of the volatility shifts. The ICSS endogenously identifies changes in volatility of stock returns. The technique is not much explored in empirical analysis of stock markets. Aggarwal et al. (1999) examined emerging stock markets in Asia and Latin America, and recently Hammoudeh and Li (in press) investigated the sudden changes in volatility for the volatile Gulf Arab stock markets.2 This paper is the first to investigate transition economies using this technique. Specifically, we investigate the emerging stock markets of the Czech Republic, Hungary, Poland, Slovakia and Slovenia over the period 1994–2006. Our empirical results indicate that a sudden change in volatility seems to arise from the evolution of emerging stock markets in an earlier period, from exchange rate policy changes and financial crises. Evidence also reveals that when sudden shifts are taken into account in the conventional GARCH models, the persistence of volatility is reduced significantly in every series. It suggests that many previous studies may have overestimated the degree of volatility persistence that exists in financial time series. The remainder of the paper is organised as follows. In Section 2, potential factors for the shifts in volatility are briefly discussed, which rationalises the usefulness of the current study. Methodology and data are described in Sections 3 and 4, respectively. In Section 5, the result of the ICSS algorithm and GARCH model are presented. Section 6 will conclude.

This paper investigates sudden shifts of volatility and re-examines volatility persistence for the stock markets of new EU members during the period 1994–2006. With the ICSS algorithm methodology, the time path of volatility in returns appears to be effectively identified for these transition economies. In this paper, sudden shifts are largely explained by domestic, economic and financial factors, and also they are more likely to be the major causes behind the shifts. This finding is consistent with that of Aggarwal et al. (1999) who find strong country-specific factors for emerging Asian and Latin American stock markets. This is contrasted with the findings of Hammoudeh and Li (in press) who find that major global events are the dominant factors for Gulf Arab stock markets. In common to these studies and also those of Ewing and Malik (2005) and Malik (2003), when the sudden shifts are incorporated in the GARCH model, the persistence of volatility has considerably reduced. It is argued that the development of stock markets for these transition economies provides a vehicle for mobilizing household savings more for equity finance hence favoring the corporate sector. This is also important for promoting international capital flows. This also enables firms to have less reliance on debt finance, reducing the risk of a credit crunch, while creating a less risky financial structure. These transition economies, therefore, face an imperative issue of establishing stable stock markets, while shifting policy towards integration with the European Union. Although high volatility is found by several shifts in the earlier period, as the integration process with the EU proceeds, sudden changes have declined, accompanied by a downward trend in volatility. The important implication is that joining the EU may be a contributory factor to stabilizing the stock markets of these transition economies.