One model with an interaction
Intermediate Regression in R
Richie Cotton
Data Evangelist at DataCamp
What is an interaction?
In the fish dataset
The effect of length on the expected mass is different for different species.
More generally
The effect of one explanatory variable on the expected response changes depending on the value of another explanatory variable.
Specifying interactions
No interactions
response ~ explntry1 + explntry2
With interactions (implicit)
response_var ~ explntry1 * explntry2
With interactions (explicit)
response ~ explntry1 + explntry2 + explntry1:explntry2
No interactions
mass_g ~ length_cm + species
With interactions (implicit)
mass_g ~ length_cm * species
With interactions (explicit)
mass_g ~ length_cm + species + length_cm:species
Running the model
lm(mass_g ~ length_cm * species, data = fish)
Call:
lm(formula = mass_g ~ length_cm * species, data = fish)
Coefficients:
(Intercept) length_cm speciesPerch speciesPike
-1035.348 54.550 416.172 -505.477
speciesRoach length_cm:speciesPerch length_cm:speciesPike length_cm:speciesRoach
705.971 -15.639 -1.355 -31.231
Easier to understand coefficients
mdl_inter <- lm(mass_g ~ species + species:length_cm + 0, data = fish)
Call:
lm(formula = mass_g ~ species + species:length_cm + 0, data = fish)
Coefficients:
speciesBream speciesPerch speciesPike speciesRoach
-1035.35 -619.18 -1540.82 -329.38
speciesBream:length_cm speciesPerch:length_cm speciesPike:length_cm speciesRoach:length_cm
54.55 38.91 53.19 23.32
Familiar numbers
speciesBream speciesPerch speciesPike speciesRoach
-1035.35 -619.18 -1540.82 -329.38
speciesBream:length_cm speciesPerch:length_cm speciesPike:length_cm speciesRoach:length_cm
54.55 38.91 53.19 23.32
coefficients(mdl_bream)
(Intercept) length_cm
-1035.34757 54.54998
Let's practice!
Intermediate Regression in R
