Erasmus university rotterdam

Table 1. Summary statistics of individual company data

Country	No. of firms	No. firms in analysis	Weekly Mean * (SD)	Weekly Mean ** (SD)	Weekly Mean*** (SD)
Egypt	118	79	.417 (7.41)	.451 (6.86)	.179 (10.52)
Israel	256	135	.085 (6.80)	.115 (6.08)	-.121(10.58)
Morocco	8	38	.293 (4.80)	.358 (4.71)	-.171 (5.35)
Turkey	357	264	.201 (8.83)	.254 (8.50)	-.174 (10.88)
Total	739	516	-	-	-

* Jan 5^th 2001 – Sept 25^th 2009 (Total period, 455 observations)

** Jan 5^th 2001 – Aug 29^th 2008 (Before crisis, 399 observations)

*** Aug 29^th 2008 – Sept 25^th 2009 (During crisis, 56 observations)

Source : Thomson Financial DataStream (all returns are adjusted for divididend payouts)

Table 2 shows the average weekly returns for the MSCI indices of the emerging markets as well as for the world market. These indices are used as proxies for the market indices. The risk-free rate, a 10- year US treasury bill is also displayed. The average returns of the individual companies included in the analysis differ from the returns on the MSCI country indices. However the main views remain the same, weekly returns are highest for the Egyptian stock market (0.49%), this would result in an annual return of over 25%. Weekly returns are lowest for Israel’s stock market (0.10%) for the total period, this would result in an annual return of approximately 5%. As was the case for the individual company returns, the standard deviation is highest for Turkey (6.99) and lowest for Morocco (2.81) for the total period. The proxy for the world market, the MSCI World displays positive weekly returns during the period before the financial crisis (0.07%) but becomes negative during the crisis (-0.27%), as expected the standard deviation of the world market is lower than the standard deviation of the individual countries. Table 2 also shows measures of skewness and kurtosis. Skewness is a measure of symmetry of the return distribution. A positive value means that the distribution is skewed to the right, while a negative value indicates that the distribution is skewed to the left. Kurtosis is a measure of peakedness of a distribution. Kurtosis measures the shape of a distribution relative to a normal distribution, the typical value of kurtosis is three. The Jarque-Bera statistic combines these two measures to determine whether a distribution is normally distributed. Table 2 shows that the return distributions for the total period are skewed to the left, thus negative. In line with this finding, Bekaert, Erb, Harvey and Viskanta (1998) find that over the period 1987-1997 the MSCI indices display negative skewness, however the IFC indices display positive skewness during this period. It must be noted that the assumption of normality of the return distributions does not hold for the markets used in this study. For all countries and periods, except Israel during the financial crisis, the assumption of normally distributed returns must be rejected. This finding is in line with earlier research on the return distributions of emerging markets. Bekaert et al. (1998) investigate skewness and kurtosis and find evidence that the distribution of emerging market returns are not-normal. Interestingly table 2 shows that the returns of the markets are closer to being normally distributed during a period of financial crisis, the Jarque-Bera measures are much smaller and for some markets the assumption of normality cannot be clearly rejected.