Dropping predictors
To obtain support for your decision, you compare the model fit of the maximal model to the model excluding the lagged coupon effect by its AIC.
You can get the AIC value by using the function AIC()
on the corresponding model object. This is easy for the maximal model which is given by the extended.model
object. To get the AIC value for the model excluding Coupon.lag
you first need to apply the function update()
on the extended.model
object. In addition to updating a model for the inclusion of an additional predictor, you can also drop a predictor. You just need to put a minus sign in front of corresponding predictor.
This exercise is part of the course
Building Response Models in R
Exercise instructions
- Get the AIC value for the
extended.model
object by using the functionAIC()
. - Update the
extended.model
object by droppingCoupon.lag
using the functionupdate()
. Get the AIC for the updated model by using the functionAIC()
.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Obtain the AIC
___(extended.model)
# Update the AIC by single term deletion
___(___(extended.model, . ~ . ___))