This project attempts to analyze the harmful impact of smoking on the chances of pre-term birth among newborns with the help of a logistic regression model. This is done by developing a model after considering the AIC and BIC step- wise and forward-built models, taking into account the accuracy and reasonableness. The model thus formed suggests that the chances of pre-term birth are significantly associated with demographics, physiology and smoking. Moreover, it finds that smoking during pregnancy may be a characteristic associated with higher pre-term births, but is not very significant and that demographic and physiological factors have stronger associations.
Data source: https://ids-702-f20.github.io/Course-Website/hw/hw-02.html, https://www.stat.berkeley.edu/users/statlabs/labs.html
Data Analysis Report