4.1 Building Regression Models
Leverage the power of correlation and regression to understand relationships and make predictions with data.
Correlation
Practice Correlation
Correlation Extremes
Practice Correlation Extremes
Simple Linear Regression
Practice Simple Linear Regression
Calculating Error
Practice Calculating Error
Regression and Prediction
Practice Regression and Prediction
Nonlinear Relationships
Practice Nonlinear Relationships
Simpson's Paradox
Use Multiple Variables
Understand Coefficients
Practice Understand Coefficients
Optimize Coefficients
Practice Optimize Coefficients
Assess the Model
Practice Assess the Model
Choose Variables
Practice Choose Variables
Avoid Overfitting
Practice Avoid Overfitting
Interpret the Model
Practice Interpret the Model
Use Categorical Variables
Practice Use Categorical Variables
Explore Interactions
Practice Explore Interactions
Transform Variables
Practice Transform Variables
Course description
This course introduces correlation and regression, which are used to quantify the strength of the relationship between variables and to compute the slope and intercept of the regression line. It explores two applications of these methods, using correlated measurements to make informed guesses for measurements that are not available, and making predictions for future events. Datasets used in these lessons include weights and other measurements from penguins and a time series of annual average temperatures. Later lesson explore nonlinear relationships and Simpson's paradox.
Topics covered
- Correlation
- Regression
- Mean squared error
- Mean absolute error
Prerequisites and next steps
Exploring Data Visually Introduction to Probability
Up next
4.2 Case Study: Maximizing Electric Car Value
This is the Case Study for Building Regression Models. In it, you will use correlation and regression to dive into data on electric vehicles to discover what a buyer gets for their money.
Jump ahead