In "T2DRecords.csv" data below (converted to jpeg in order to attach to this question), five independent variables are used to predict the probability that a patient has type 2 diabetes. The independents are:

BodyWeight:Each patient's body weight reported in the survey.

GymVisits:Each patient's number of visits to gyms every week.

State:In which state each patient is current residing in. The assumption is certain local diet may increase the risk of getting type 2 diabetes.

SurgarIntake:The average amount of sugar each patient takes every week.

ChLevel:Each patient'scholesterol level measured in this survey.

The dependent variable in this data set isT2Dwith 1 indicating a patient has type 2 diabetes and 0 indicating a patient does not have type 2 diabetes.

Questions:

1. Build a logistic regression model, and use all given independent variables to predict the probability that a patient has type 2 diabetes. In your assignment submission, please make a screen capture of your analysis results. Please indicate which independent variables are significant, and the practical implications if they are significant.

2. Build a decision tree model, and use all given independent variables to predict the probability that a patient has type 2 diabetes. In your assignment submission, please make a screen capture of your analysis results. Also, depict the data in a decision tree diagram to visualize your analysis.

3. Use the ROC curve to compare which model is more accurate in prediction based on the current data set. Please make screen captures of the ROC curves you build.

To access "T2DRecords.csv" please click: https://docs.google.com/spreadsheets/d/1mL7XyvDU9Kjr3i5-YD4mIJmiy5-9oFrJ8cuC5DMk3sg/edit#gid=0