Logistic Regression in a Nutshell

• WHAT IS LOGISTIC REGRESSION?

Logistic Regression is a model that uses a logistic function to model a dependent variable. It make predictions about the probability of a particular event occurring. It estimates the parameters of a logistic model and estimates it between 0 and 1. Logistic Regression also deals with classification issues.

• WHEN TO USE LOGISTIC REGRESSION

You would want to use logistic regression instead of linear regression when you are dealing with binary data. This is when the dependent variable is categorical, for example: yes or no and pass or fail. You would also want the independent variable to be either discrete (ordinal), discrete (nominal), or continuous. An example of this would be temperature, weight, on a scale, or something like color.

• REAL WORLD EXAMPLES

Logistic regression can be used in many different professional circumstances. A few real world examples of when to use logistic regression are: predicting if an incoming email is spam or not, predicting if a credit card transaction is fraudulent or not, predicting whether a tumor is benign or malignant, predicting if a people will buy a particular products or not, and predicting whether a student will complete their course on time or not.

• LOGISTIC REGRESSION VS. LINEAR REGRESSION

Here are the key differences between logistic and linear regression. Linear regression requires a linear relationship between the dependent and independent variables while logistic regression does not. In linear regression the data is modeled using a straight line while in logistic regression he probability of the event is represented as a linear function of a combination of predictor variables. In logistic regression the independent variables should not be correlated with one another but in linear regression the independent variables could be correlated with each other. Finally, linear regression models continuous numerical data while logistic regression models binary data.