Module 2: Multiple Regression

In this module, you will learn about multiple regression. You'll explore how to model relationships with multiple predictor variables, conduct t-tests to determine variable significance, and compare model fit using adjusted R-squared.

Objective 1: Model with multiple independent features

Learn how to build regression models that incorporate multiple predictor variables to explain a single outcome variable.

• Understanding the concept of multiple linear regression
• Interpreting coefficients in a multiple regression model
• Building and evaluating multiple regression models
• Understanding the benefits of using multiple predictors

Objective 2: Use the t-test results to determine feature significance

Learn how to assess the statistical significance of individual predictor variables in a multiple regression model.

• Understanding the t-test for individual coefficients
• Interpreting p-values for predictors
• Determining which variables are statistically significant
• Making decisions about variable inclusion based on significance

Objective 3: Compare model fit using adjusted r-squared

Learn how to use adjusted R-squared to compare the fit of different regression models.

• Understanding the limitations of regular R-squared
• Calculating and interpreting adjusted R-squared
• Using adjusted R-squared to compare models with different numbers of predictors
• Making model selection decisions based on adjusted R-squared

Multiple Regression

Resources:

The notebook for this guided project is DS_132_Multiple_Regression.ipynb in the GitHub repository.

Multiple Regression Assignment

In this module assignment, found in the file DS_132_Multiple_Regression_Assignment_AG.ipynb in the GitHub repository, you'll apply your knowledge of multiple regression to a real-world dataset:

Tasks:

1. Build a multiple regression model with several predictor variables
2. Conduct t-tests to determine variable significance
3. Compare model fit using R-squared