Using Nonparametric Models to Determine the Most Important Factors Affecting the Weight of the Newborn

You are currently viewing Using Nonparametric Models to Determine the Most Important Factors Affecting the Weight of the Newborn

Using Nonparametric Models to Determine the Most Important Factors Affecting the Weight of the Newborn

Using Nonparametric Models to Determine the Most Important Factors Affecting the Weight of the Newborn

A Thesis Submitted to

Council of  The Administration and Economics/ Karbala University as Partial fulfillment of the Requirements for the Degree of Master of Science in Statistics

Presented by researcher

Hawraa A Hussain Al khazali

Supervised By

Prof. Dr. Jassim N Hussain

Nonparametric models are used in this thesis, which are polynomial regression models and spline models, which are part of the nonparametric statistics. They differ from parametric models, as they do not depend on previous assumptions, but depend mainly on the data, and they are called freely distributed models because their parameters are flexible and fixed. The weight of the newborn is one of the important things to know about the possibility of the growth of the newborn, because abnormal weights are associated with negative effects on the health of the child, and it affects the mental abilities and others. It is necessary to know this because it is related to his health condition, and the main problem is the phenomenon of abnormal weight in newborns so resorted The researcher aims to study the factors that affect the weight of the newborn, where five factors have been studied (mother’s age, number of births, pregnancy duration, chronic diseases, number of visits to health care centers).

 The researcher used the non-parametric methods above, and the factors that affected the child’s weight at birth were known, as well as knowing the best model that gave results among the models used in the applied side using the criteria of differentiation, the mean square error MSE, the Akaiki criterion AIC and the Biz informational criterion BIC, and through the study it was found that the best model among The models that were applied was a quadratic polynomial model because it gave the lowest values ​​for the comparison criteria. According to this model, the factors affecting the child’s weight were arranged, and the most influential factor was (chronic diseases), followed by (the number of births), the duration of pregnancy, and finally the number of visits to health centers.