This paper's main motivation is to examine the seasonal pattern of overall mortality of children whose ages are below five years in Egypt. We used the data of vital statistics for births and deaths, which have been collected by the Central Agency for Public Mobilization and Statistics (CAPMAS) from January 2002 to December 2010 in Egypt. Seasonal effects on under–five and infant mortality were measured by several different statistical models such as the classic multiple regression model (CMR), Poisson regression model (PR), generalized additive models (GAM) and neural network models (NNM). During the study period, all children below 5 years in the study population and at the aggregate level all cases of deaths of whom have been recorded. There were significant seasonal variations of under–five mortality. The mortality risk was significantly high in January and August compared with other months. This paper demonstrates that overall mortality of children under–five of age in Egypt has a seasonal pattern. Generally during the months of January and August of each year during the period of study, the death rate for the under–five children increases. This evidence, which was obtained here, may help officials to intervene to reduce the mortality rate among the under–five children in Egypt. The neural networks and the generalized additive models are the optimal alternative methods to estimate the parameters with less bias for detecting seasonality of child mortality data.