Babies with low birth weight have some health problems in later years. Therefore, it is important to estimate before the birth whether a new born baby will have a low birth weight or not. In order to obtain this estimation, logistic regression model is a suitable choice. Logistic regression analysis is a modelling technique which is used when the dependent variable is categorical. It is also easily interpreted. When the dependent variable has only two categories, the logistic regression is called binary logistic regression. Logistic regression has parametric and nonparametric solutions. MARS method is a nonparametric method which can be used as an alternative to the parametric solutions in the analysis of logistic regression. The nonparametric models require fewer assumptions compared to the parametric ones and they are also more flexible. In the application, a binary logistic regression model has been fitted to estimate whether a new born baby will have a low birth weight or not. The model has been estimated based on the MARS method. In the analysis, data belonging to 982 subjects have been investigated by applying the MARS software. In the conclusion part, the findings are interpreted.
Low birth weight Logistic regression Maximum likelihood MARS
Düşük doğum ağırlığı Lojistik regresyon En çok olabilirlik MARS
Birincil Dil | Türkçe |
---|---|
Konular | İstatistik |
Bölüm | Araştırma Makaleleri |
Yazarlar | |
Yayımlanma Tarihi | 15 Temmuz 2013 |
Yayımlandığı Sayı | Yıl 2013 Cilt: 10 Sayı: 2 |