Question

Provide an examplewhere unsupervised learning (categorical response) would be used as the analysis. Describe the predictor(s), the response variable, and explain whether prediction or inference would be of primary interest.

Answer 1

I use Logistic regression when my response variable is categorical in nature. but for the application of Logistic regression it is required that the response variable has only two categories. suppose I want to predict the probability of default for a credit card company. here the dependent variable is type of default which can be either yes (1) or no (0). The independent variable are the income and age. The equation of Logistic regression in this case is given by: P = e^(bo+b1*x1+b2*x2)/(1+e^(bo+b1*x1+b2*x2)). beta1 indicates the change in Probability for default with a unit increase in income. beta to indicate the change in probability of default with 1 year increase in age.