Logistic regression (LR) is a statistical method for binary classification.

There are four key concepts that are central to understanding LR:

• Odds and Log Odds: These concepts form the basis of logistic regression, helping us understand how to transform probabilities into a more manageable form.
• Probability Estimation: This is the process of calculating the likelihood of an outcome occurring, which is central to making predictions in logistic regression.
• Sigmoid Function: The mathematical function that transforms linear predictions into probability values between 0 and 1, making it essential for binary classification tasks.
• Binary vs. Multinomial Logistic Regression: Understanding the differences between these two types of logistic regression and when to apply each one.

You can explore each of these concepts in detail by following the links below.

Content map

To learn more about key concepts in logistic regression, read each of the following sections:

Odds and Log Odds

Probability Estimation

Sigmoid Function

Binary vs. Multinomial Logistic Regression

Next element

Model fitting