Factors Associated with Birthweight: An Application of the Multiple Skew-Normal Regression

In the present research, we consider birthweight as a skew-normal random variable. It is possible the birthweight relative risk factors change under the skew-normal assumption. Our results confirm that the data for groups of babies, LBW (low birthweight), normal and combined groups (p<0.001) are skew. It is to be noted that the relevant factors associated with birthweight for two groups of babies, LBW and normal babies, are quite different and we should not analyze them together. We also found out that the relevant risk factors for birthweight are different in the normal and skew-normal models. Such that some relevant variables that are not significant in normal assumption become significant in skew-normal and versa as. Moreover the association magnitude of risk factors in the normal and skew-normal is different.