Calculating the LOO estimate
Now let's practice using the loo
package on our Spotify model so that we can determine which model provides the best fit to our data. The songs
dataset is already loaded.
This exercise is part of the course
Bayesian Regression Modeling with rstanarm
Exercise instructions
- Estimate a model predicting
popularity
fromsong_age
- Print the LOO approximation for this model with 1 predictor
- Estimate a model predicting
popularity
fromsong_age
,artist_name
, and their interaction - Print the LOO approximation for this model with 2 independent variables
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Estimate the model with 1 predictor
model_1pred <- ___(___ ~ ___, data = songs)
# Print the LOO estimate for the 1 predictor model
___(model_1pred)
# Estimate the model with both predictors
model_2pred <- ___(___ ~ ___ * ___, data = songs)
# Print the LOO estimates for the 2 predictor model
___(model_2pred)