Part 4 | General Linear Model
// build simple models of relationships //Part 4 introduces the workhorse of empirical economics: regression analysis. In Part 3, we used regression with a single predictor to test whether patterns in our data were real or just noise. But economic relationships are rarely so simple. The General Linear Model not only lets us test simple hypotheses, but also lets us handle multiple predictors of different types - continuous, categorical, and their interactions - while maintaining our core tools of visualization and residual analysis. These tools are the foundation for modern science.
Part 4.1 ~ Extending the t-Test
The two-sample t-Test is a simple extension of the one-sample t-Test that tests for differences in means.
Concept 4.1 // Extending the t-Test
A simple extension of the t-test
Exercise 4.1 // Interactive t-Test Extensions
Interactive examples extending basic t-tests
Homework 4.1
Due on Friday at 5PM on Gradescope
Part 4.2 ~ Simple Linear Models
The t-Test is a simple type of linear regression model. Use simple linear regression to build more complex relationships between variables. Interpret coefficients and p-values. Assess model assumptions using residual plots.
Concept 4.2 // Simple Linear Models
The workhorse statistical model in science
Exercise 4.2 // Interactive Linear Models
Interactive examples with linear regression
Homework 4.2
Due on Friday at 5PM on Gradescope
Part 4.3 ~ Indicator Variables
Use indicator variables to include multiple intercepts by category. Interpret indicator coefficients and p-values.
Concept 4.3 // Indicator Variables
A binary indicator variable takes the values of 1 and 0.
Exercise 4.3 // Neighborhood Data Analysis
Using indicator variables with neighborhood datasets
Homework 4.3
Due on Friday at 5PM on Gradescope
Part 4.4 ~ Statistical Controls
Build multiple regression models with several predictors. Interpret partial regression coefficients and p-values.
Concept 4.4 // Statistical Controls
Use multivariate regressions for statistical controls
Exercise 4.4 // USA Census Data
Statistical controls with real census data
Homework 4.4
Due on Friday at 5PM on Gradescope
MiniExams
Part 4 MiniExam tests your understanding of linear modeling concepts including t-tests, regression analysis, indicator variables, and statistical controls.
MiniExam 06
Try solving the demo then check against the video.