A simulation-based approach to the study of coefficient of variation of dividend yields

Existing empirical studies of dividend yields and dividend policies either make no assumption or the normal distribution of the dividend yields data. The statistical results will be biased because they cannot reflect the finite support set property of dividend yields which can only range from 0 to 1. We posit that the assumption that dividend yields follow a beta distribution is more appropriate. The coefficient of variation (CV) is used to measure the stability of dividend yields. If we assume dividend yields follow a normal distribution, then the maximum likelihood estimate for coefficient of variation is given by sx¯. This only gives us a point estimate, which cannot depict the full picture of the sampling distribution of the coefficient of variation. A simulation-based approach is adopted to estimate CV under the beta distribution. This approach will give us a point estimate as well as the empirical sampling distribution of CV. With this approach, we study the stability of dividend yields of the Hang Seng index and its sub-indexes of the Hong Kong stock market and compare the results with the traditional approach.