STAT 331 Case #3 Logistic Regression

 

Each group will submit ONE WORD copy on Blackboard by the deadline indicated on the syllabus (also on Blackboard), with subject: STAT331 HW#-group#. Please include a cover page with all your group members.

 

1. Logistic regression & Decision tree

Please find the dataset named hw3.csv on bblearn.

 

Random sample a training data set that contains 90% of the original data points.  

 

(i) Start with descriptive summary as usual. Please select some variables (2 or 3) and report their boxplots, histograms, and scatter plots.

(ii) Conduct logistic regression on the training data using all the variables (To better interpret regression results, do not normalize data). 

(iii) Conduct classification tree analysis on the training data and show the tree plot with text (Please note that the cost function is unknown in this case because we have limited domain knowledge of this dataset).  

(iv) Test the out-of-sample performance (testing set) for the logistic regression  model from (ii) and the classification tree model from (iii). Please make a comparison using ROC curve. Which one is better?

.

 

Write a brief report including all labeled figures and tables.