Stock Return Volatility on Emerging Eastern European Markets

Abstract

A common ¢nding for developed stock markets is that negative shocks entering the market lead to a larger return volatility than positive shocks of a similar magnitude. We consider two emerging Eastern European markets where the ¢rst point of investigation is whether an analogous asymmetric characteristic is re£ected in emerging markets. The second point of investigation is whether the ¢ndings di¡er depending on the institutional microstructure of the exchange being examined. Hence, econometric techniques are adjusted and a``doublecensored Tobit GARCH'' model is developed. We ¢nd that no asymmetry exists on either emerging market but asymmetry does exist for a stock return series on the developed market; possible reasons for this are proposed.

The impact of speculative trading activities on the speculative market: a case of Taiwan stock index futures market

“…Hence GARCH(1, 1) is found to be capable of capturing the heteroscedasticity of conditional volatility. This finding is consistent with previous results (Bollerslev et al, 1992;Shields, 1997), therefore GARCH(1, 1) model is employed for further analysis.…”

The comovements of stock markets in Hungary, Poland and the Czech Republic

“…The results in Table 5 show that the symmetric conditional variance of the GARCH model is justified because the tests never reject the null hypothesis. The absence of asymmetric effects in Eastern Europe is also documented by Shields (1997). This study discusses a number of explanations for the symmetric behavior of volatility, such as differences in investor behavior, the particularities of the transition process or a slower spreading of information.…”

“…Among the small number of existing articles Emerson et al (1996) model Bulgarian stocks. Shields (1997) compares GARCH specifications for returns from the Warsaw and Budapest stock exchanges. Scheicher (1999) analyses volatilities and jumps in Polish stock returns.…”

Predictability of stock markets with disequilibrium trading

“…The data used are for returns, on a session to session basis, with two sessions a week in 1991 and 1992, three sessions a week since the beginning of 1993, four sessions a week from mid-1994, and finally five sessions a week from the end of 1994. (Detailed descriptive and econometric analyses of the Warsaw Stock Exchange, including an analysis of possible calendar effects, can be found in Shields (1997Shields ( ,1999, in addition to Gordon and Rittenberg (1995), Bolt and Milobedzki (1994a,b)). We do not use direct quantitative information concerning identification of the disequlibrium trading sessions.…”