Follow

How can I create an RMA (reduced major axis) fit curve in a Grapher script?

Grapher's fit curves use the least squares fit method. Grapher does not have an automatic way to create the reduced major axis regression equation directly. However, RMA and least squares fit curves are related. There is a script available that does calculate the RMA equation and add it to your existing graph as a function plot. That script is attached to this article.


The downside to this method is that there is not a reported R2 value for the fit. However, some literature says that the R2 value is the same for the RMA fit as it is for the least squares fit method. So, if you create a line/scatter plot and add the Linear fit, the statistics for the linear fit are the same as the statistics for the RMA fit. You can then run the script to add the RMA fit curve to the graph.

To add an RMA (reduced major axis) fit to your graph:

  1. Create a 2D scatter plot from the data by clicking the Graphs | Create | Basic | Scatter Plot (Grapher 12 Scatter Plot) command.
  2. Click the Developer | User Scripts | Create RMA fit curve command. The script runs and the RMA curve is added to the graph.
  3. To get the R2 statistic, click on the line/scatter plot in the Object Manager.
  4. In the Property Manager, click on the Plot tab.
  5. Click the Add/Edit button next to the Fits command.
  6. In the Fits dialog, select Linear, Y=B*X+A.
  7. Click Add.
  8. Click OK.
  9. Click on the Fit 1: Linear fit curve in the Object Manager.
  10. In the Property Manager, click on the Display button next to the Report option. The statistics appear in a report window.
  11. Highlight the Coefficient of determination, R-sq'd value (R2) value in the report.
  12. Click Edit | Copy in the report window.
  13. In the main plot window, click Home | Clipboard | Paste. The value is pasted to the center of the plot window. You can click on the value and move it to the desired location or edit the text.

 

Updated September 22, 2016

Was this article helpful?
0 out of 0 found this helpful
Have more questions? Submit a request

0 Comments

Please sign in to leave a comment.