Question

Provide two classification scenarios where you can use classification tree models but not logistic regression models. For each problem, identify a target variable and four possible predictor variables.

Answer 1

Both classification tree model and  regression models are able to handle continuous and categorical data just fine. It is actually difficult to predict which technique will be better. It really depends on the specific problem and the data we are having.

Logistic regression model is actually searching for a single linear decision boundary in your feature space, whereas a decision tree model is essentially partitioning your feature space into half-spaces decision boundaries. The net effect is that you have a non-linear more than one decision boundary.

Lastly, another thing to consider is that decision tree model can automatically take into account interactions between variables. With logistic regression, you'll have to manually add those interactions by  yourself.

It is  always a good idea to just try both models and do cross-validation. This will help you find out which one is more likely and suitable.

Example1: In sample of 30 students with three variables Gender (Boy/ Girl), Class (IX/ X) and Height (5 to 6 ft). 15 out of 30 play cricket in free time. Now, if we want to create a model to predict which students will be playing cricket during leisure period? In this problem, we need to segregate students who play cricket in their leisure time based on input variable among all three.

This is where decision tree model helps. It will segregate the students based on all values of three variable and identify the variable, which creates the best homogeneous sets of students.In this the target variable can be "Student will play cricket or not” either  YES or NO. All the variable that will be used to predict the target variable is a predictor variable.

Example 2: Simple decision tree for buying a car, The predictor variables could include milage, colour , space etc and the target variable will be whether to bur a car or not.

In simple terms we can conclude that predictor variable is a variable whose values will be used to predict the value of the target variable.