Logistic regression is a statistical method for predicting binary outcomes like acceptance/rejection using predictor variables. It yields a predicted probability of a positive outcome. Customers with high predicted probabilities from a logistic model of credit card acceptance may be targeted for offers since they are more likely to accept. Similarly, a logistic model of credit scores found that odds of a good score increase with higher net cash flow and experience, while odds of a bad score increase with higher debt ratios.
Logistic regression is a statistical method for predicting binary outcomes like acceptance/rejection using predictor variables. It yields a predicted probability of a positive outcome. Customers with high predicted probabilities from a logistic model of credit card acceptance may be targeted for offers since they are more likely to accept. Similarly, a logistic model of credit scores found that odds of a good score increase with higher net cash flow and experience, while odds of a bad score increase with higher debt ratios.
Logistic regression is a statistical method for predicting binary outcomes like acceptance/rejection using predictor variables. It yields a predicted probability of a positive outcome. Customers with high predicted probabilities from a logistic model of credit card acceptance may be targeted for offers since they are more likely to accept. Similarly, a logistic model of credit scores found that odds of a good score increase with higher net cash flow and experience, while odds of a bad score increase with higher debt ratios.
Logistic regression is a statistical method for predicting binary outcomes like acceptance/rejection using predictor variables. It yields a predicted probability of a positive outcome. Customers with high predicted probabilities from a logistic model of credit card acceptance may be targeted for offers since they are more likely to accept. Similarly, a logistic model of credit scores found that odds of a good score increase with higher net cash flow and experience, while odds of a bad score increase with higher debt ratios.
• Method of choice when the dependent variable is binary and
assumes only two discrete values • By inputting values for the predictor variables for each new customer – the logistic model will yield a predicted probability • Customers with high ‘predicted probabilities’ may be chosen to receive an offer since they seem more likely to respond positively EXAMPLE: Credit Card Offering
• Dependent Variable - whether the customer signed up for a gold card
offer • Predictor Variables - other bank services the customer used plus financial and demographic customer information • By inputting values for the predictor variables for each new customer, the logistic model will yield a predicted probability • Customers with high ‘predicted probabilities’ may be chosen to receive the offer since they seem more likely to respond positively Credit Score Modelling
• It can be seen that the most predictive variables
are net cash flow, client's business experience and total debt ratio. Looking at the regression coefficient shows that the odds of the company of being bad are increasing with the increase of total debt ratio. • On the contrary, odds of the company of being good increases with the increase of the net cash flow