Implications of market frictions: serial correlations in indexes on the emerging stock markets in Central and Eastern Europe

This paper focuses on the implications of market frictions in the context of serial correlations in indexes on the Central and Eastern European (CEE) stock markets. Market frictions, such as non-trading effects, bid/ask spreads, other transaction costs, etc., may be detected by direct measurement, or by indirect identification. Direct measurement of frictions is difficult as intraday trading data are unavailable in the case of most of the emerging CEE stock markets. Indirect identification may be conducted by detecting some empirical phenomena. One of them is evidence of serial correlations in indexes, the so-called Fisher effect. We explore the problem of serial correlations in indexes on the eight CEE stock markets using data samples from each CEE market separately, as well as a ‘common trading window’ approach, which is widely applied in the case of databases with multivariate time series. The evidence is that ‘nonsynchronous trading effect II’ between markets may substantially disrupt the analysis of index returns on a domestic market. Using a synchronized database, one may erroneously conclude that Fisher effect does not exist, although it is present.