Additional Example of a Logistic Plot

This example uses the Car Physical Data.jmp sample data table to show an additional example of a logistic plot. Suppose you want to use weight to predict car size (Type) for 116 cars. Car size can be one of the following, from smallest to largest: Sporty, Small, Compact, Medium, or Large.

Select Help > Sample Data Library and open Car Physical Data.jmp.

In the Columns panel, right-click on the icon to the left of Type, and select Ordinal.

Right-click on Type, and select Column Info.

From the Column Properties menu, select Value Ordering.

Move the data in the following top-down order: Sporty, Small, Compact, Medium, Large.

Click OK.

Select Analyze > Fit Y by X.

Select Type and click Y, Response.

Select Weight and click X, Factor.

10.

Click OK.

The report window appears.

Example of Type by Weight Logistic Plot

In Example of Type by Weight Logistic Plot, note the following observations:

•

The first (bottom) curve represents the probability that a car at a given weight is Sporty.

•

The second curve represents the probability that a car is Small or Sporty. Looking only at the distance between the first and second curves corresponds to the probability of being Small.

•

As you might expect, heavier cars are more likely to be Large.

•

Markers for the data are drawn at their x-coordinate, with the y position jittered randomly within the range corresponding to the response category for that row.

If the x -variable has no effect on the response, then the fitted lines are horizontal and the probabilities are constant for each response across the continuous factor range. Examples of Sample Data Table and Logistic Plot Showing No y by x Relationship shows a logistic plot where Weight is not useful for predicting Type.

Examples of Sample Data Table and Logistic Plot Showing No y by x Relationship

Note: To re-create the plots in Examples of Sample Data Table and Logistic Plot Showing No y by x Relationship and Examples of Sample Data Table and Logistic Plot Showing an Almost Perfect y by x Relationship, you must first create the data tables shown here, and then perform steps 7-10 at the beginning of this section.

If the response is completely predicted by the value of the factor, then the logistic curves are effectively vertical. The prediction of a response is near certain (the probability is almost 1) at each of the factor levels. Examples of Sample Data Table and Logistic Plot Showing an Almost Perfect y by x Relationship shows a logistic plot where Weight almost perfectly predicts Type.

Note: In this case, the parameter estimates become very large and are marked unstable in the regression report.

Examples of Sample Data Table and Logistic Plot Showing an Almost Perfect y by x Relationship