O'Reilly logo

Practical Predictive Analytics by Ralph Winters

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

Comparing AIC among the candidate models

As mentioned in earlier chapters, AIC is a metric that you can use to help you select a model. The plotty='T' option also produced a plot of the AICs for the top models found. In the function call, we indicate that we wanted to see up to five.

According to the author of the package, a reasonable rule of thumb is to consider models that fall below the horizontal red line.

According to that definition, that would eliminate models 4 and 5, and leave models 1, 2, and 3 to consider.

Here is the plot of the AICs for the top five models:

 

We can also specify a variable importance plot by using the plot statement ...

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

Start Free Trial

No credit card required