(Mapping of Infant Mortality Cases in Peninsular Malaysia using Empirical Bayes Method)

The objective of this study was to map the infant mortality cases over Peninsular Malaysia by district for the period of 1991 to 2000. Relative risks estimation based on empirical Bayes method was used in this study. Three methods of estimation were described which include moment method, maximum likelihood method and combination of moment and maximum likelihood method. The robustness of the parameters estimation was examined using Bootstrap method. The study indicated that the gap between the low risk areas and the high risk areas are larger in the early decade of 2000 compared to the early 1990s eventhough the infant mortality rate is declining at the national level. The east coast areas of the Peninsular Malaysia still remain in the high risk category over the period of the study. Moreover, the maps obtained indicated the occurrence of clustering effect in the infant mortality risk. Based on the Bootstrap method, all parameters estimation obtained in this study were robust.