Erasmus university rotterdam

Investigation of the data

The first part of the empirical investigation required an investigation of the available data. A summary of the descriptive statistics of the stock market data is available in table 1. The return data on all companies that were taken into account in this study are sampled at a weekly frequency. The returns were adjusted for dividends, splits and share issues. The lognormal returns were calculated using the following formula:
[8]

When dealing with emerging stock markets it is necessary to determine whether a stock is active or inactive. Since emerging stock markets are often illiquid and dominated by small number of stocks.

Only stocks that are actively traded and regularly traded are considered for inclusion in the analysis. In order to correct for thin trading only stocks that traded at least 90% (α = 0.90) of the weeks in the period available and had at least 100 observations were included in the analysis, except for the period during the financial crisis where all stocks had 56 observations. Furthermore, to filter the data for errors all returns that exceeded a weekly return of 100% were removed from the sample. Table 1 shows the number of firms that were included in the analysis. Table 1 also shows that the historical average returns for the four emerging countries are positive during the total period, however returns are negative during the financial crisis. As expected the standard deviations of these countries are high and increase in the period from August 2008 to September 2009. The Egyptian companies display the highest average weekly returns during the entire period (0.417%), and even during the period of the financial crisis returns remain positive (0.179%). However, the high returns are accompanied by high risk. Zhang and Wihlborg (2004) find a monthly average return for Turkey of 6.6% (sd = 19.299%).