This study tests whether daily returns series of Gulf Cooperation Council (GCC) Stock Markets are an
approximation of normal distribution or not. The assumption of normal distribution is common and crucial
for many statistical tools used by the stock market researchers and analysts. Saudi, Qatar, Kuwait and
Oman stock market indices are examined in this study. Based on Chi-square, Kolmogorov-Smirnov test,
Autocorrelation Function and Partial Autocorrelation Functions, this study does not accept the null
hypothesis that the distribution of the market returns is not different from the normal distribution.
The two prime considerations of the investors are the return and the risk. Return
represents the expected relative accrual on the value of investment. Risk is the expected
variance of the expected return. Any investor perceives a return as an inducement for the
postponement of consumption and as a compensation for the risk associated with the
investment. Valuation of assets depends upon the associated return and risk. Thus, the
investment decisions are made based on the expected return and risk. Formulation of a
probability distribution of ‘one period’ returns guides the analyst in arriving at a fair value
of an asset and usually the historical relative frequency distribution of returns acts as an
objective and close approximation of the future probability distribution. Even when we
estimate a subjective future probability distribution the historical distribution of returns
forms the foundation. The estimated probability distribution of returns facilitates the
analyst in making quantitative estimates of expected return and risk associated with the
securities.
There are two components that constitute the return on a security, namely, the
dividend and the price difference. The price difference is the most predominant
determinant of the security’s return since the market is said to be instantaneously
impounding the information, including the dividend information, into the pricing
process. The ‘one-period’ return is computed either as a percentage change or as a
logarithmic price change during the period. In the context of securities research the latter
measure appears to be the most preferred for certain analytical advantages. Risk is
measured as the dispersion of the probability distribution of ‘one-period’ returns. It is the
variance of returns around the expected returns. Notational presentation of the expected
return and risk calculations is as follows: |
