O'Reilly logo

Industrial Statistics with Minitab by Xavier Tort-Martorell Llabres, Lluis Marco Almagro, Pere Grima Cintas

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

23.5 Selection of the Best Equation: Best Subsets

Stat > Regression > Best Subsets

It generates all possible equations, together with some goodness of fit measures for each one of them. This allows us to select and thoroughly study (based for instance on significance of coefficients, residual analysis) those that seem to be more appropriate.

In the following, the square of the variable Power, named Pow2, is entered as a new explanatory variable.

nc23uf010.eps

The list that appears, with all default options (the best two equations with one to four explanatory variables), is:

nc23uf011.eps
The selection of the best equation depends also on aspects such as the simplicity of the model or its ease of interpretation. In many cases, it is questionable to say that we have the ‘best’ model. Thus, perhaps it would be better to talk about a ‘good’ model.

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