Introduction to Regression with statsmodels in Python

Overview

Predict housing prices and ad click-through rate by implementing, analyzing, and interpreting regression analysis with statsmodels in Python.

Linear regression and logistic regression are two of the most widely used statistical models. They act like master keys, unlocking the secrets hidden in your data. In this course, you’ll gain the skills you need to fit simple linear and logistic regressions. Through hands-on exercises, you’ll explore the relationships between variables in real-world datasets, including motor insurance claims, Taiwan house prices, fish sizes, and more. By the end of this course, you’ll know how to make predictions from your data, quantify model performance, and diagnose problems with model fit.

Syllabus

Simple Linear Regression Modeling

You’ll learn the basics of this popular statistical model, what regression is, and how linear and logistic regressions differ. You’ll then learn how to fit simple linear regression models with numeric and categorical explanatory variables, and how to describe the relationship between the response and explanatory variables using model coefficients.

Predictions and model objects

In this chapter, you’ll discover how to use linear regression models to make predictions on Taiwanese house prices and Facebook advert clicks. You’ll also grow your regression skills as you get hands-on with model objects, understand the concept of "regression to the mean", and learn how to transform variables in a dataset.

Assessing model fit

In this chapter, you’ll learn how to ask questions of your model to assess fit. You’ll learn how to quantify how well a linear regression model fits, diagnose model problems using visualizations, and understand each observation's leverage and influence to create the model.

Simple Logistic Regression Modeling

Learn to fit logistic regression models. Using real-world data, you’ll predict the likelihood of a customer closing their bank account as probabilities of success and odds ratios, and quantify model performance using confusion matrices.