Explain whether each scenario is a classification or regression problem, and ind

Question

Explain whether each scenario is a classification or regression problem, and indicate whether we are most interested in inference or prediction and why. Finally, provide n (number of observations) and p (number of predictors).

1. The mortgage division of a bank collects the following information of their customers: mortgage amount, property value, property type, interest rate, annual income, age, and number of late payments. The division wants to use the information to categorize their 30,000 customers into three categories: high, medium, and low risk of loan default.

Explanation / Answer

This is a clear case of a classification problem as we are trying to segment our customer population into 3 types of customers: High, medium and low risk of loan default. Therefore we are classifying them into 3 types. Clearly here the dependent variable that is the Customer category is a categorical variable which means it is a classification problem. Regression problems on the other hand have a numerical dependent variable.

Here we are not really predicting any future output for a set of predictors but we are inferencing out that based on the given inputs what would be the customer types. therefore it is an inference type of a problem.

The predictors given in the problem are:

Mortgage amount

Property Value

Property type

Interest Rate

Annual Income

Age

Number of late payments

Therefore there are 7 predictors used in the given problem to classify 30,000 customers.

Therefore p = 7 and n= 30,000

Related Questions

Navigate

• Browse (All)
• Browse E
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.