Interaction Term

Also known as · Interaction Terms

The product of two regressors, as in yi=β0+β1xi+β2 malei+β3(xi⋅malei)+uiy_i = \beta_0 + \beta_1 x_i + \beta_2\,male_i + \beta_3 (x_i\cdot male_i) + u_i. The dummy β2\beta_2 shifts the intercept while the interaction β3\beta_3 changes the slope, so the effect of xx can differ by group.

When to use

Use it to give each group its own slope, to write the Chow Test as a pooled regression, or to estimate group-specific Elasticity values (for example by beverage type).

See Lecture 11.

Appears in