GitHub - mjoneil21/Homework-5-R: R portion of Homework 5: This homework assignment featured both logistic regression and tree based regression (simple decision tree, random forest, ect.) After a quick look at the data through the str and summary function I constructed my logrithmic model. The model outputs a value between 0 and 1 however, the data we are predicting is binary, 0 or 1. Therefore, a rule was needed to change whatever values the model spat out into o's and 1's. A good starting point is .5 and above is 1 and below .5 is 0. After that, I moved on to the simple decision tree, bootstrap aggregation and lastly random forest. Because my train data was so much larger than my test data my models ended up being far more accurate on the test dataset than on the train. Another cool trick I learned is that when using the random forest package you can order the variables in the model from most important to least important based on mean decrease gini coeficient.

mjoneil21 / Homework-5-R Public

Notifications You must be signed in to change notification settings
Fork 0
Star 0

R portion of Homework 5: This homework assignment featured both logistic regression and tree based regression (simple decision tree, random forest, ect.) After a quick look at the data through the str and summary function I constructed my logrithmic model. The model outputs a value between 0 and 1 however, the data we are predicting is binary, 0…

0 stars 0 forks Branches Tags Activity

Star

Notifications

Branches Tags

Name		Name	Last commit message	Last commit date
Latest commit History 10 Commits
.gitignore		.gitignore
HW5_prob3_ojsales.Rmd		HW5_prob3_ojsales.Rmd
HW5_prob3_ojsales.html		HW5_prob3_ojsales.html
OJ.Rproj		OJ.Rproj