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

23.4 Option Buttons

23.4.1 Graphs

The following graphs allow us, through the analysis of residuals, to assess the validity of the fitted model.

nc23uf005.eps

There are different types of residuals:

Regular: Raw residuals, computed as the difference between the true value and the corresponding fitted value.
Standardized: Computed as the regular residual divided by the standard deviation of the residuals.
Deleted: Similar to the Standardized, but in each case the data point whose residual is to be computed is excluded in the estimation of the coefficients of the model.

The requested graphs are Four in one, which displays in one graphical window the four more relevant residual plots and the graph of residuals versus the variable Power, since as previously illustrated, it seems that the relationship between power and maximum speed is not linear but quadratic.

nc23uf006.eps

The shape of the plot of residuals versus fitted values (cloud of points in a parabolic shape) suggests the possibility of improving the model by transforming a variable. The plot of the residuals versus the variable Power suggests the convenience of a transformation of this variable: for instance taking the square of it.

nc23uf007.eps

23.4.2 Options ...

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