Question

An insurance company has past data that will allow them to build a model to predict if a newly filed insurance claim is fraudulent. The outcome variable is "fraud" (yes or no), and there is a combination of quantitative and categorical predictor variables. Which are appropriate methods that can be used to predict which newly filed insurance claims are fraudulent?

A multiple linear regression model is appropriate for this data, but logistic regression is not

A logistic regression model is appropriate for this data, but multiple linear regression is not

Both logistic regression model and a multiple linear regression models are appropriate methods for this data

Neither a logistic regression model nor a multiple linear regression model is appropriate for this data

Answer 1

Q. filed insurance claim is fraudulent. The outcome variable is "fraud" (yes or no), and there is a combination of quantitative and categorical predictor variables. Which are appropriate methods that can be used to predict which newly filed insurance claims are fraudulent?

Answer:-

Because predictor is binary we will use logistic regression we cannot use linear regression.we can make dummy variables for categorical data.

A logistic regression model is appropriate for this data, but multiple linear regression is not.