Font Size: 

A new mortality model based on a mixture distribution function is proposed. We mix a half-normal distribution with a generalization of the skew-normal distribution. As a result we get a six-paramters distribution function having a good fit with a wide variety of mortality patterns. This model is fitted to several mortality data and compared with the Siler model (five parameters). The main feature of the proposed model is that it has a relatively good fit even in mortality pattern with a high "accident" hump (this happens, for instance, in countries undergoing armed conflicts or with a high HIV prevalence), while the Siler model cannot catch this hump.