The similar iterative process in problem 1 has been applied on the cancer dataset to build a logistic regression with biomarkers. It went through two iterations (EXCEL FILE BELOW FROM GOOGLE DOCS)

https://docs.google.com/spreadsheets/d/1azNRtK7d593jjOlh1gFmrC8c2EFJVVPW9pZS8NxfdPo/edit?usp=sharing

a. In your final model, what factors have positive influence (i.e., positive slope) on risk of cancer and what factors have negative influence (negative slope) on risk of cancer?

b. use your final model to predict the probability of cancer for this patient with the biomarker profile below:

AFP CEA CA125 CA199 CA50

5.05 10.7 8.12 20.62 57.39