1.
Open the SAT.jmp sample data table.
2.
Select Analyze > Fit Y by X.
3.
Select 2004 Verbal and click Y, Response.
4.
Select % Taking (2004) and click X, Factor.
5.
Example of SAT Scores by Percent Taking
6.
From the red triangle menu for Bivariate Fit, select Fit Special. The Specify Transformation or Constraint window appears. For a description of this window, see Description of the Specify Transformation or Constraint Window.
The Specify Transformation or Constraint Window
7.
Within Y Transformation, select Natural Logarithm: log(y).
8.
9.
Example of Fit Special Report
Example of Fit Special Report shows the fitted line plotted on the original scale. The model appears to fit the data well, as the plotted line goes through the cloud of points.
1.
Open the Big Class.jmp sample data table.
2.
Select Analyze > Distribution.
3.
Select height and weight and click Y, Columns.
4.
5.
1.
From the Big Class.jmp sample data table, select Analyze > Fit Y by X.
2.
Select Std weight and click Y, Response.
3.
Select Std height and click X, Factor.
4.
6.
From the red triangle menu, select Fit Orthogonal. Then select each of the following:
Example of Orthogonal Fitting Options
The scatterplot in Example of Orthogonal Fitting Options shows the standardized height and weight values with various line fits that illustrate the behavior of the orthogonal variance ratio selections. The standard linear regression (Fit Line) occurs when the variance of the X variable is considered to be very small. Fit X to Y is the opposite extreme, when the variation of the Y variable is ignored. All other lines fall between these two extremes and shift as the variance ratio changes. As the variance ratio increases, the variation in the Y response dominates and the slope of the fitted line shifts closer to the Y by X fit. Likewise, when you decrease the ratio, the slope of the line shifts closer to the X by Y fit.
The data in the Weight Measurements.jmp sample data table shows the height and weight measurements taken by 40 students.
1.
Open the Weight Measurements.jmp sample data table.
2.
Select Analyze > Fit Y by X.
3.
Select weight and click Y, Response.
4.
Select height and click X, Factor.
5.
Example of Robust Fit
This example uses the Hot Dogs.jmp sample data table. The Type column identifies three different types of hot dogs: beef, meat, or poultry. You want to group the three types of hot dogs according to their cost variables.
1.
Open the Hot Dogs.jmp sample data table.
2.
Select Analyze > Fit Y by X.
3.
Select $/oz and click Y, Response.
4.
Select $/lb Protein and click X, Factor.
5.
8.
Click OK. If you look at the Group By option again, you see it has a check mark next to it.
9.
To color the points according to Type, proceed as follows:
11.
Select Type in the column list and click OK.
Example of Group By
The ellipses in Example of Group By show clearly how the different types of hot dogs cluster with respect to the cost variables.
1.
Open the Big Class.jmp sample data table.
2.
Select Analyze > Fit Y by X.
3.
Select weight and click Y, Response.
4.
Select height and click X, Factor.
5.
6.
Select Fit Line from the red triangle menu.
10.
11.
Select Fit Line from the red triangle menu.
Example of Regression Analysis for Whole Sample and Grouped Sample
The scatterplot to the left in Example of Regression Analysis for Whole Sample and Grouped Sample has a single regression line that relates weight to height. The scatterplot to the right shows separate regression lines for males and females.